Wing-shaped walls: A directional effect of obstacles on manual avoidance

Visual information can be used to plan, start, and coordinate manual movements in obstacle avoidance. An intriguing example of visuomotor coordination is the effect of wing-shaped walls, in which walls are oriented away from or toward a moving agent. A historical story from medieval Japan recounts that wing-shaped walls disrupted the agent's movement more when oriented toward the agent than when oriented away from the agent. This study aimed at examining whether the disruptive effect of wing-shaped walls occurs in a schematic situation represented on a 2D plane. In this study, we conducted psychophysical experiments in which participants were asked to move a stylus from a start point to a goal while avoiding multiple line obstacles that were arranged alternately at a course. In the two experiments, we manipulated the orientation and the size of the visible parts of the obstacles systematically. We found that the obstacles oriented toward the agent produced frequent contacts with the agent and attracted manual movements to the endpoints of obstacles. We discussed possible interpretations of the results in the context of attentional guidance.

Improved Hybrid Model for Obstacle Detection and Avoidance in Robot Operating System Framework (Rapidly Exploring Random Tree and Dynamic Windows Approach)

The integration of machine learning and robotics brings promising potential to tackle the application challenges of mobile robot navigation in industries. The real-world environment is highly dynamic and unpredictable, with increasing necessities for efficiency and safety. This demands a multi-faceted approach that combines advanced sensing, robust obstacle detection, and avoidance mechanisms for an effective robot navigation experience. While hybrid methods with default robot operating system (ROS) navigation stack have demonstrated significant results, their performance in real time and highly dynamic environments remains a challenge. These environments are characterized by continuously changing conditions, which can impact the precision of obstacle detection systems and efficient avoidance control decision-making processes. In response to these challenges, this paper presents a novel solution that combines a rapidly exploring random tree (RRT)-integrated ROS navigation stack and a pre-trained YOLOv7 object detection model to enhance the capability of the developed work on the NAV-YOLO system. The proposed approach leveraged the high accuracy of YOLOv7 obstacle detection and the efficient path-planning capabilities of RRT and dynamic windows approach (DWA) to improve the navigation performance of mobile robots in real-world complex and dynamically changing settings. Extensive simulation and real-world robot platform experiments were conducted to evaluate the efficiency of the proposed solution. The result demonstrated a high-level obstacle avoidance capability, ensuring the safety and efficiency of mobile robot navigation operations in aviation environments.

Motor flexibility to stabilize the toe position during obstacle crossing in older adults: an investigation using an uncontrolled manifold analysis

Introduction

An age-related decrease in the ability to exploit the abundant degrees of freedom of the body, referred to as motor flexibility, leads to a heightened fall risk. The present study investigated motor flexibility to stabilize the toe position during obstacle crossing in older adults and its correlation with the magnitude of foot elevation.

Methods

Twenty-six older adults (70.9 ± 7.4 years old) and 21 younger adults (25.4 ± 5.0 years old) walked and crossed an obstacle, during which the dominant limb was always the leading limb. An uncontrolled manifold (UCM) analysis was used to quantify the flexibility during obstacle crossing as the synergy index, with the vertical toe position being regarded as the performance variable and the segment angles of the lower limbs as the elemental variables.

Results and discussion

The results showed that older participants had a significantly lower synergy index for the trailing limb before the moment of obstacle crossing than younger participants, suggesting reduced flexibility in part. The results also showed that, regardless of age, foot elevation was negatively correlated with the synergy index, suggesting that a so-called "conservative strategy" (i.e., a tendency to show extraordinarily high foot elevation to ensure collision avoidance) may be related to their reduced motor flexibility.

Analysis of collision avoidance in honeybee flight

Insects are excellent at flying in dense vegetation and navigating through other complex spatial environments. This study investigates the strategies used by honeybees (Apis mellifera) to avoid collisions with an obstacle encountered frontally during flight. Bees were trained to fly through a tunnel that contained a solitary vertically oriented cylindrical obstacle placed along the midline. Flight trajectories of bees were recorded for six conditions in which the diameter of the obstructing cylinder was systematically varied from 25 mm to 160 mm. Analysis of salient events during the bees' flight, such as the deceleration before the obstacle, and the initiation of the deviation in flight path to avoid collisions, revealed a strategy for obstacle avoidance that is based on the relative retinal expansion velocity generated by the obstacle when the bee is on a collision course. We find that a quantitative model, featuring a controller that extracts specific visual cues from the frontal visual field, provides an accurate characterization of the geometry and the dynamics of the manoeuvres adopted by honeybees to avoid collisions. This study paves the way for the design of unmanned aerial systems, by identifying the visual cues that are used by honeybees for performing robust obstacle avoidance flight.

Toward the Intelligent, Safe Exploration of a Biomimetic Underwater Robot: Modeling, Planning, and Control

Safe, underwater exploration in the ocean is a challenging task due to the complex environment, which often contains areas with dense coral reefs, uneven terrain, or many obstacles. To address this issue, an intelligent underwater exploration framework of a biomimetic robot is proposed in this paper, including an obstacle avoidance model, motion planner, and yaw controller. Firstly, with the aid of the onboard distance sensors in robotic fish, the obstacle detection model is established. On this basis, two types of obstacles, i.e., rectangular and circular, are considered, followed by the obstacle collision model's construction. Secondly, a deep reinforcement learning method is adopted to plan the plane motion, and the performances of different training setups are investigated. Thirdly, a backstepping method is applied to derive the yaw control law, in which a sigmoid function-based transition method is employed to smooth the planning output. Finally, a series of simulations are carried out to verify the effectiveness of the proposed method. The obtained results indicate that the biomimetic robot can not only achieve intelligent motion planning but also accomplish yaw control with obstacle avoidance, offering a valuable solution for underwater operation in the ocean.

A Review of Sensing Technologies for Indoor Autonomous Mobile Robots

As a fundamental issue in robotics academia and industry, indoor autonomous mobile robots (AMRs) have been extensively studied. For AMRs, it is crucial to obtain information about their working environment and themselves, which can be realized through sensors and the extraction of corresponding information from the measurements of these sensors. The application of sensing technologies can enable mobile robots to perform localization, mapping, target or obstacle recognition, and motion tasks, etc. This paper reviews sensing technologies for autonomous mobile robots in indoor scenes. The benefits and potential problems of using a single sensor in application are analyzed and compared, and the basic principles and popular algorithms used in processing these sensor data are introduced. In addition, some mainstream technologies of multi-sensor fusion are introduced. Finally, this paper discusses the future development trends in the sensing technology for autonomous mobile robots in indoor scenes, as well as the challenges in the practical application environments.

An Unmanned Aerial Vehicle Indoor Low-Computation Navigation Method Based on Vision and Deep Learning

Recently, unmanned aerial vehicles (UAVs) have found extensive indoor applications. In numerous indoor UAV scenarios, navigation paths remain consistent. While many indoor positioning methods offer excellent precision, they often demand significant costs and computational resources. Furthermore, such high functionality can be superfluous for these applications. To address this issue, we present a cost-effective, computationally efficient solution for path following and obstacle avoidance. The UAV employs a down-looking camera for path following and a front-looking camera for obstacle avoidance. This paper refines the carrot casing algorithm for line tracking and introduces our novel line-fitting path-following algorithm (LFPF). Both algorithms competently manage indoor path-following tasks within a constrained field of view. However, the LFPF is superior at adapting to light variations and maintaining a consistent flight speed, maintaining its error margin within ±40 cm in real flight scenarios. For obstacle avoidance, we utilize depth images and YOLOv4-tiny to detect obstacles, subsequently implementing suitable avoidance strategies based on the type and proximity of these obstacles. Real-world tests indicated minimal computational demands, enabling the Nvidia Jetson Nano, an entry-level computing platform, to operate at 23 FPS.

Outdoor Navigation Assistive System Based on Robust and Real-Time Visual-Auditory Substitution Approach

Blindness affects millions of people worldwide, leading to difficulties in daily travel and a loss of independence due to a lack of spatial information. This article proposes a new navigation aid to help people with severe blindness reach their destination. Blind people are guided by a short 3D spatialised sound that indicates the target point to follow. This sound is combined with other sonified information on potential obstacles in the vicinity. The proposed system is based on inertial sensors, GPS data, and the cartographic knowledge of pedestrian paths to define the trajectory. In addition, visual clues are used to refine the trajectory with ground floor information and obstacle information using a camera to provide 3D spatial information. The proposed method is based on a deep learning approach. The different neural networks used in this approach are evaluated on datasets that regroup navigations from pedestrians' point-of-view. This method achieves low latency and real-time processing without relying on remote connections, instead using a low-power embedded GPU target and a multithreaded approach for video processing, sound generation, and acquisition. This system could significantly improve the quality of life and autonomy of blind people, allowing them to reliably and efficiently navigate in their environment.

Intelligent Head-Mounted Obstacle Avoidance Wearable for the Blind and Visually Impaired

Individuals who are Blind and Visually Impaired (BVI) take significant risks and dangers on obstacles, particularly when they are unaccompanied. We propose an intelligent head-mount device to assist BVI people with this challenge. The objective of this study is to develop a computationally efficient mechanism that can effectively detect obstacles in real time and provide warnings. The learned model aims to be both reliable and compact so that it can be integrated into a wearable device with a small size. Additionally, it should be capable of handling natural head turns, which can generally impact the accuracy of readings from the device's sensors. Over thirty models with different hyper-parameters were explored and their key metrics were compared to identify the most suitable model that strikes a balance between accuracy and real-time performance. Our study demonstrates the feasibility of a highly efficient wearable device that can assist BVI individuals in avoiding obstacles with a high level of accuracy.

Approaching Process in Walking through an Aperture for Individuals with Stroke

Muroi et al. show that individuals with stroke have improved collision avoidance behavior when passing through an aperture while entering from the paretic-side of the body. However, the underlying mechanism remains unknown. We reanalyzed Muroi et al.'s data to reveal how individuals with stroke walk through an aperture by examining changes in walking velocity and behavioral complexity (i.e., sample entropy, an index of (ir)regularity of time series, regarded lower entropy as more regular and less complex) by focusing on the approaching process. The results showed that individuals with stroke reduced their walking velocity and behavioral complexity before passing through the narrow aperture when approaching from the paretic side. We interpreted that the improved obstacle avoidance when penetrating from the paretic side may be due to careful body rotation and adjusting the walking velocity in advance.

Characterizing neurocognitive impairments in Parkinson's disease with mobile EEG when walking and stepping over obstacles

The neural correlates that help us understand the challenges that Parkinson's patients face when negotiating their environment remain under-researched. This deficit in knowledge reflects the methodological constraints of traditional neuroimaging techniques, which include the need to remain still. As a result, much of our understanding of motor disorders is still based on animal models. Daily life challenges such as tripping and falling over obstacles represent one of the main causes of hospitalization for individuals with Parkinson's disease. Here, we report the neural correlates of naturalistic ambulatory obstacle avoidance in Parkinson's disease patients using mobile EEG. We examined 14 medicated patients with Parkinson's disease and 17 neurotypical control participants. Brain activity was recorded while participants walked freely, and while they walked and adjusted their gait to step over expected obstacles (preset adjustment) or unexpected obstacles (online adjustment) displayed on the floor. EEG analysis revealed attenuated cortical activity in Parkinson's patients compared to neurotypical participants in theta (4-7 Hz) and beta (13-35 Hz) frequency bands. The theta power increase when planning an online adjustment to step over unexpected obstacles was reduced in Parkinson's patients compared to neurotypical participants, indicating impaired proactive cognitive control of walking that updates the online action plan when unexpected changes occur in the environment. Impaired action planning processes were further evident in Parkinson's disease patients' diminished beta power suppression when preparing motor adaptation to step over obstacles, regardless of the expectation manipulation, compared to when walking freely. In addition, deficits in reactive control mechanisms in Parkinson's disease compared to neurotypical participants were evident from an attenuated beta rebound signal after crossing an obstacle. Reduced modulation in the theta frequency band in the resetting phase across conditions also suggests a deficit in the evaluation of action outcomes in Parkinson's disease. Taken together, the neural markers of cognitive control of walking observed in Parkinson's disease reveal a pervasive deficit of motor-cognitive control, involving impairments in the proactive and reactive strategies used to avoid obstacles while walking. As such, this study identified neural markers of the motor deficits in Parkinson's disease and revealed patients' difficulties in adapting movements both before and after avoiding obstacles in their path.

Application of Path Planning and Obstacle Avoidance for Riverbank Inspection

Most coastal trash comes from land. To prevent and control ocean pollution, it should be handled using sources that can maintain a clean ocean and improve the marine ecological environment. The proposed system can be used to inspect riverbanks and identify garbage on riverbanks. This waste can then be cleaned up before flowing into the sea. In this study, we utilized an unmanned aerial vehicle (UAV) to inspect riverbanks and applied path planning and obstacle avoidance to enhance the efficiency of UAV performance and ensure good adaptability in a complicated environment. Since most rivers in the middle and upper sections of the study area are rough and meandering, path planning was first addressed so that the drone could use the shortest path and less energy to perform the inspection task. Branches frequently protrude from the riverbank on both sides. Therefore, an instant obstacle avoidance algorithm was added to avoid various obstacles. Path planning was based on an Improved Particle Swarm Optimization (IPSO). A fuzzy system was added to the IPSO to adjust the parameters that could shorten the planned path. The Artificial Potential Field (APF) was applied for real-time dynamic obstacle avoidance. The proposed UAV system could be used to perform riverbank inspection successfully.

A Dynamic Path-Planning Method for Obstacle Avoidance Based on the Driving Safety Field

Establishing an accurate and computationally efficient model for driving risk assessment, considering the influence of vehicle motion state and kinematic characteristics on path planning, is crucial for generating safe, comfortable, and easily trackable obstacle avoidance paths. To address this topic, this paper proposes a novel dual-layered dynamic path-planning method for obstacle avoidance based on the driving safety field (DSF). The contributions of the proposed approach lie in its ability to address the challenges of accurately modeling driving risk, efficient path smoothing and adaptability to vehicle kinematic characteristics, and providing collision-free, curvature-continuous, and adaptable obstacle avoidance paths. In the upper layer, a comprehensive driving safety field is constructed, composed of a potential field generated by static obstacles, a kinetic field generated by dynamic obstacles, a potential field generated by lane boundaries, and a driving field generated by the target position. By analyzing the virtual field forces exerted on the ego vehicle within the comprehensive driving safety field, the resultant force direction is utilized as guidance for the vehicle's forward motion. This generates an initial obstacle avoidance path that satisfies the vehicle's kinematic and dynamic constraints. In the lower layer, the problem of path smoothing is transformed into a standard quadratic programming (QP) form. By optimizing discrete waypoints and fitting polynomial curves, a curvature-continuous and smooth path is obtained. Simulation results demonstrate that our proposed path-planning algorithm outperforms the method based on the improved artificial potential field (APF). It not only generates collision-free and curvature-continuous paths but also significantly reduces parameters such as path curvature (reduced by 62.29% to 87.32%), curvature variation rate, and heading angle (reduced by 34.11% to 72.06%). Furthermore, our algorithm dynamically adjusts the starting position of the obstacle avoidance maneuver based on the vehicle's motion state. As the relative velocity between the ego vehicle and the obstacle vehicle increases, the starting position of the obstacle avoidance path is adjusted accordingly, enabling the proactive avoidance of stationary or moving single and multiple obstacles. The proposed method satisfies the requirements of obstacle avoidance safety, comfort, and stability for intelligent vehicles in complex environments.

Trajectory Tracking and Obstacle Avoidance of Robotic Fish Based on Nonlinear Model Predictive Control

The attainment of accurate motion control for robotic fish inside intricate underwater environments continues to be a substantial obstacle within the realm of underwater robotics. This paper presents a proposed algorithm for trajectory tracking and obstacle avoidance planning in robotic fish, utilizing nonlinear model predictive control (NMPC). This methodology facilitates the implementation of optimization-based control in real-time, utilizing the present state and environmental data to effectively regulate the movements of the robotic fish with a high degree of agility. To begin with, a dynamic model of the robotic fish, incorporating accelerations, is formulated inside the framework of the world coordinate system. The last step involves providing a detailed explanation of the NMPC algorithm and developing obstacle avoidance and objective functions for the fish in water. This will enable the design of an NMPC controller that incorporates control restrictions. In order to assess the efficacy of the proposed approach, a comparative analysis is conducted between the NMPC algorithm and the pure pursuit (PP) algorithm in terms of trajectory tracking. This comparison serves to affirm the accuracy of the NMPC algorithm in effectively tracking trajectories. Moreover, a comparative analysis between the NMPC algorithm and the dynamic window approach (DWA) method in the context of obstacle avoidance planning highlights the superior resilience of the NMPC algorithm in this domain. The proposed strategy, which utilizes NMPC, demonstrates a viable alternative for achieving precise trajectory tracking and efficient obstacle avoidance planning in the context of robotic fish motion control within intricate surroundings. This method exhibits considerable potential for practical implementation and future application.

Application of Deep Reinforcement Learning to UAV Swarming for Ground Surveillance

This paper summarizes in depth the state of the art of aerial swarms, covering both classical and new reinforcement-learning-based approaches for their management. Then, it proposes a hybrid AI system, integrating deep reinforcement learning in a multi-agent centralized swarm architecture. The proposed system is tailored to perform surveillance of a specific area, searching and tracking ground targets, for security and law enforcement applications. The swarm is governed by a central swarm controller responsible for distributing different search and tracking tasks among the cooperating UAVs. Each UAV agent is then controlled by a collection of cooperative sub-agents, whose behaviors have been trained using different deep reinforcement learning models, tailored for the different task types proposed by the swarm controller. More specifically, proximal policy optimization (PPO) algorithms were used to train the agents' behavior. In addition, several metrics to assess the performance of the swarm in this application were defined. The results obtained through simulation show that our system searches the operation area effectively, acquires the targets in a reasonable time, and is capable of tracking them continuously and consistently.

A decoupled Bayesian method for snake robot control in unstructured environment

This paper presents a method which avoids the common practice of using a complex coupled snake robot model and performing kinematic analysis for control in cluttered environments. Instead, we introduce a completely decoupled dynamical Bayesian formulation with respect to interacted snake robot links and environmental objects, which requires much lower complexity for efficient and robust control. When a snake robot does not interact with obstacles, it runs by a simple serpenoid controller. However, when it exhibits interaction with environments, defined as close proximity or collision with targets and/or obstacles, we extend the conventional Bayesian framework by modeling such interactions in terms of stimuli. The proposed 'multi-neural-stimulus function' represents the cumulative effect of both external environmental influences and internal constraints of the snake robot. It implicitly handles the 'unexpected collision' problem and thus solve the difficult data association and shape adjustment problems for snake robot control in an innovative way. Preliminary experimental results have demonstrated promising performance of the proposed method comparing with the state-of-the-art.

Corrigendum: Ring attractor bio-inspired neural network for social robot navigation

[This corrects the article DOI: 10.3389/fnbot.2023.1211570.].

Ring attractor bio-inspired neural network for social robot navigation

Introduction

We introduce a bio-inspired navigation system for a robot to guide a social agent to a target location while avoiding static and dynamic obstacles. Robot navigation can be accomplished through a model of ring attractor neural networks. This connectivity pattern between neurons enables the generation of stable activity patterns that can represent continuous variables such as heading direction or position. The integration of sensory representation, decision-making, and motor control through ring attractor networks offers a biologically-inspired approach to navigation in complex environments.

Methods

The navigation system is divided into perception, planning, and control stages. Our approach is compared to the widely-used Social Force Model and Rapidly Exploring Random Tree Star methods using the Social Individual Index and Relative Motion Index as metrics in simulated experiments. We created a virtual scenario of a pedestrian area with various obstacles and dynamic agents.

Results

The results obtained in our experiments demonstrate the effectiveness of this architecture in guiding a social agent while avoiding obstacles, and the metrics used for evaluating the system indicate that our proposal outperforms the widely used Social Force Model.

Discussion

Our approach points to improving safety and comfort specifically for human-robot interactions. By integrating the Social Individual Index and Relative Motion Index, this approach considers both social comfort and collision avoidance features, resulting in better human-robot interactions in a crowded environment.

A Framework for Optimal Navigation in Situations of Localization Uncertainty

The basic functions of an autonomous vehicle typically involve navigating from one point to another in the world by following a reference path and analyzing the traversability along this path to avoid potential obstacles. What happens when the vehicle is subject to uncertainties in its localization? All its capabilities, whether path following or obstacle avoidance, are affected by this uncertainty, and stopping the vehicle becomes the safest solution. In this work, we propose a framework that optimally combines path following and obstacle avoidance while keeping these two objectives independent, ensuring that the limitations of one do not affect the other. Absolute localization uncertainty only has an impact on path following, and in no way affects obstacle avoidance, which is performed in the robot's local reference frame. Therefore, it is possible to navigate with or without prior information, without being affected by position uncertainty during obstacle avoidance maneuvers. We conducted tests on an EZ10 shuttle in the PAVIN experimental platform to validate our approach. These experimental results show that our approach achieves satisfactory performance, making it a promising solution for collision-free navigation applications for mobile robots even when localization is not accurate.

Utilizing Drone-Based Ground-Penetrating Radar for Crime Investigations in Localizing and Identifying Clandestine Graves

The decomposition of a body is influenced by burial conditions, making it crucial to understand the impact of different conditions for accurate grave detection. Geophysical techniques using drones have gained popularity in locating clandestine graves, offering non-invasive methods for detecting surface and subsurface irregularities. Ground-penetrating radar (GPR) is an effective technology for identifying potential grave locations without disturbance. This research aimed to prototype a drone system integrating GPR to assist in grave localization and to develop software for data management. Initial experiments compared GPR with other technologies, demonstrating its valuable applicability. It is suitable for various decomposition stages and soil types, although certain soil compositions have limitations. The research used the DJI M600 Pro drone and a drone-based GPR system enhanced by the real-time kinematic (RTK) global positioning system (GPS) for precision and autonomy. Tests with simulated graves and cadavers validated the system's performance, evaluating optimal altitude, speed, and obstacle avoidance techniques. Furthermore, global and local planning algorithms ensured efficient and obstacle-free flight paths. The results highlighted the potential of the drone-based GPR system in locating clandestine graves while minimizing disturbance, contributing to the development of effective tools for forensic investigations and crime scene analysis.

Shape of an obstacle affects the mediolateral trajectory of the lower limb during the crossing process

In previous studies involving obstacle crossing, vertical foot clearance has been used as an indicator of the risk of contact. Under normal circumstances, individuals do not always cross over obstacles with the same height on both sides, and depending on the shape of the obstacle, the risk of contact may differ depending on the foot elevation position. Therefore, we investigated whether task-related control of the mediolateral foot position is adapted to the shape of the obstacle. Sixteen healthy young adults performed a task in which they crossed over two obstacles with different shapes while walking: a trapezoidal obstacle and a rectangular obstacle, as viewed from the frontal plane. It was shown that when crossing over a trapezoidal obstacle, the participants maintained foot clearance by controlling the mediolateral direction, which chose the height that needed to be cleared. The results of this study suggest that the lower limb movements that occur during obstacle crossing are controlled not only in the vertical direction but also in the mediolateral direction by adjusting the foot trajectory to reduce the risk of contact. It was demonstrated that control was not only based on the height of the obstacle directly under the foot but also in the foot mediolateral direction, considering the shape of the entire obstacle, including the opposite limb.

A Cooperative Hunting Method for Multi-USV Based on the A* Algorithm in an Environment with Obstacles

A single unmanned surface combatant (USV) has poor mission execution capability, so the cooperation of multiple unmanned surface ships is widely used. Cooperative hunting is an important aspect of multi USV collaborative research. Therefore, this paper proposed a cooperative hunting method for multi-USV based on the A* algorithm in an environment with obstacles. First, based on the traditional A* algorithm, a path smoothing method based on USV minimum turning radius is proposed. At the same time, the post order traversal recursive algorithm in the binary tree method is used to replace the enumeration algorithm to obtain the optimal path, which improves the efficiency of the A* algorithm. Second, a biomimetic multi USV swarm collaborative hunting method is proposed. Multiple USV clusters simulate the hunting strategy of lions to pre-form on the target's path, so multiple USV clusters do not require manual formation. During the hunting process, the formation of multiple USV groups is adjusted to limit the movement and turning of the target, thereby reducing the range of activity of the target and improving the effectiveness of the algorithm. To verify the effectiveness of the algorithm, two sets of simulation experiments were conducted. The results show that the algorithm has good performance in path planning and target search.

Obstacle avoidance in aerial pursuit

Pursuing prey through clutter is a complex and risky activity requiring integration of guidance subsystems for obstacle avoidance and target pursuit. The unobstructed pursuit trajectories of Harris' hawks Parabuteo unicinctus are well modeled by a mixed guidance law feeding back target deviation angle and line-of-sight rate. Here we ask how their pursuit behavior is modified in response to obstacles, using high-speed motion capture to reconstruct flight trajectories recorded during obstructed pursuit of maneuvering targets. We find that Harris' hawks use the same mixed guidance law during obstructed pursuit but appear to superpose a discrete bias command that resets their flight direction to aim at a clearance of approximately one wing length from an upcoming obstacle as they reach some threshold distance from it. Combining a feedback command in response to target motion with a feedforward command in response to upcoming obstacles provides an effective means of prioritizing obstacle avoidance while remaining locked-on to a target. We therefore anticipate that a similar mechanism may be used in terrestrial and aquatic pursuit. The same biased guidance law could also be used for obstacle avoidance in drones designed to intercept other drones in clutter, or to navigate between fixed waypoints in urban environments.

IRDC-Net: Lightweight Semantic Segmentation Network Based on Monocular Camera for Mobile Robot Navigation

Computer vision plays a significant role in mobile robot navigation due to the wealth of information extracted from digital images. Mobile robots localize and move to the intended destination based on the captured images. Due to the complexity of the environment, obstacle avoidance still requires a complex sensor system with a high computational efficiency requirement. This study offers a real-time solution to the problem of extracting corridor scenes from a single image using a lightweight semantic segmentation model integrating with the quantization technique to reduce the numerous training parameters and computational costs. The proposed model consists of an FCN as the decoder and MobilenetV2 as the decoder (with multi-scale fusion). This combination allows us to significantly minimize computation time while achieving high precision. Moreover, in this study, we also propose to use the Balance Cross-Entropy loss function to handle diverse datasets, especially those with class imbalances and to integrate a number of techniques, for example, the Adam optimizer and Gaussian filters, to enhance segmentation performance. The results demonstrate that our model can outperform baselines across different datasets. Moreover, when being applied to practical experiments with a real mobile robot, the proposed model's performance is still consistent, supporting the optimal path planning, allowing the mobile robot to efficiently and effectively avoid the obstacles.

Microrobot Path Planning Based on the Multi-Module DWA Method in Crossing Dense Obstacle Scenario

A hard issue in the field of microrobots is path planning in complicated situations with dense obstacle distribution. Although the Dynamic Window Approach (DWA) is a good obstacle avoidance planning algorithm, it struggles to adapt to complex situations and has a low success rate when planning in densely populated obstacle locations. This paper suggests a multi-module enhanced DWA (MEDWA) obstacle avoidance planning algorithm to address the aforementioned issues. An obstacle-dense area judgment approach is initially presented by combining Mahalanobis distance, Frobenius norm, and covariance matrix on the basis of a multi-obstacle coverage model. Second, MEDWA is a hybrid of enhanced DWA (EDWA) algorithms in non-dense areas with a class of two-dimensional analytic vector field methods developed in dense areas. The vector field methods are used instead of the DWA algorithms with poor planning performance in dense areas, which greatly improves the passing ability of microrobots over dense obstacles. The core of EDWA is to extend the new navigation function by modifying the original evaluation function and dynamically adjusting the weights of the trajectory evaluation function in different modules using the improved immune algorithm (IIA), thus improving the adaptability of the algorithm to different scenarios and achieving trajectory optimization. Finally, two scenarios with different obstacle-dense area locations were constructed to test the proposed method 1000 times, and the performance of the algorithm was verified in terms of step number, trajectory length, heading angle deviation, and path deviation. The findings indicate that the method has a smaller planning deviation and that the length of the trajectory and the number of steps can both be reduced by about 15%. This improves the ability of the microrobot to pass through obstacle-dense areas while successfully preventing the phenomenon of microrobots going around or even colliding with obstacles outside of dense areas.

Can Genetic Algorithms Be Used for Real-Time Obstacle Avoidance for LiDAR-Equipped Mobile Robots?

Despite significant progress in robot hardware, the number of mobile robots deployed in public spaces remains low. One of the challenges hindering a wider deployment is that even if a robot can build a map of the environment, for instance through the use of LiDAR sensors, it also needs to calculate, in real time, a smooth trajectory that avoids both static and mobile obstacles. Considering this scenario, in this paper we investigate whether genetic algorithms can play a role in real-time obstacle avoidance. Historically, the typical use of genetic algorithms was in offline optimization. To investigate whether an online, real-time deployment is possible, we create a family of algorithms called GAVO that combines genetic algorithms with the velocity obstacle model. Through a series of experiments, we show that a carefully chosen chromosome representation and parametrization can achieve real-time performance on the obstacle avoidance problem.

Training for walking through an opening improves collision avoidance behavior in subacute patients with stroke: a randomized controlled trial

PURPOSE

Paretic side collisions frequently occur in stroke patients, especially while walking through narrow spaces. We determined whether training for walking through an opening (T-WTO) while entering from the paretic side would improve collision avoidance behavior and prevent falls after 6 months.

MATERIALS AND METHODS

Thirty-eight adults with moderate-to-mild hemiparetic gait after stroke who were hospitalized in a rehabilitation setting were randomly allocated to the T-WTO (n = 20) or regular rehabilitation (R-Control; n = 18) program. Both groups received five sessions of 40 min per week, for three weeks total. T-WTO included walking through openings of various widths while rotating with the paretic side in front, and R-Control involved normal walking without body rotation. Obstacle avoidance ability, 10-m walking test, timed Up and Go test, Berg Balance Scale, Activities-specific Balance Confidence, the perceptual judgment of passability, and fall incidence were assessed.

RESULTS

Collision rate and time to passage of the opening in obstacle avoidance task significantly improved in the T-WTO group compared with those in the R-Control group. Contrast, T-WTO did not lead to significant improvements in other outcomes.

CONCLUSIONS

T-WTO improved efficiency and safety in managing subacute stroke patients. Such training could improve patient outcomes/safety because of the paretic body side during walking.

CLINICAL TRIAL REGISTRATION NO.

R000038375 UMIN000033926.

Trajectory Planning of Autonomous Underwater Vehicles Based on Gauss Pseudospectral Method

This paper aims to address the obstacle avoidance problem of autonomous underwater vehicles (AUVs) in complex environments by proposing a trajectory planning method based on the Gauss pseudospectral method (GPM). According to the kinematics and dynamics constraints, and the obstacle avoidance requirement in AUV navigation, a multi-constraint trajectory planning model is established. The model takes energy consumption and sailing time as optimization objectives. The optimal control problem is transformed into a nonlinear programming problem by the GPM. The trajectory satisfying the optimization objective can be obtained by solving the problem with a sequential quadratic programming (SQP) algorithm. For the optimization of calculation parameters, the cubic spline interpolation method is proposed to generate initial value. Finally, through comparison with the linear fitting method, the rapidity of the solution of the cubic spline interpolation method is verified. The simulation results show that the cubic spline interpolation method improves the operation performance by 49.35% compared with the linear fitting method, which verifies the effectiveness of the cubic spline interpolation method in solving the optimal control problem.

Apollo: Adaptive Polar Lattice-Based Local Obstacle Avoidance and Motion Planning for Automated Vehicles

The motion planning module is the core module of the automated vehicle software system, which plays a key role in connecting its preceding element, i.e., the sensing module, and its following element, i.e., the control module. The design of an adaptive polar lattice-based local obstacle avoidance (APOLLO) algorithm proposed in this paper takes full account of the characteristics of the vehicle's sensing and control systems. The core of our approach mainly consists of three phases, i.e., the adaptive polar lattice-based local search space design, the collision-free path generation and the path smoothing. By adjusting a few parameters, the algorithm can be adapted to different driving environments and different kinds of vehicle chassis. Simulations show that the proposed method owns strong environmental adaptability and low computation complexity.

Research into Autonomous Vehicles Following and Obstacle Avoidance Based on Deep Reinforcement Learning Method under Map Constraints

Compared with traditional rule-based algorithms, deep reinforcement learning methods in autonomous driving are able to reduce the response time of vehicles to the driving environment and fully exploit the advantages of autopilot. Nowadays, autonomous vehicles mainly drive on urban roads and are constrained by some map elements such as lane boundaries, lane driving rules, and lane center lines. In this paper, a deep reinforcement learning approach seriously considering map elements is proposed to deal with the autonomous driving issues of vehicles following and obstacle avoidance. When the deep reinforcement learning method is modeled, an obstacle representation method is proposed to represent the external obstacle information required by the ego vehicle input, aiming to address the problem that the number and state of external obstacles are not fixed.

(A)symmetry during gait initiation in people with Parkinson's disease: A motor and cortical activity exploratory study

Background

Gait asymmetry and deficits in gait initiation (GI) are among the most disabling symptoms in people with Parkinson's disease (PwPD). Understanding if PwPD with reduced asymmetry during GI have higher asymmetry in cortical activity may provide support for an adaptive mechanism to improve GI, particularly in the presence of an obstacle.

Objective

This study quantified the asymmetry of anticipatory postural adjustments (APAs), stepping parameters and cortical activity during GI, and tested if the presence of an obstacle regulates asymmetry in PwPD.

Methods

Sixteen PwPD and 16 control group (CG) performed 20-trials in two conditions: unobstructed and obstructed GI with right and left limbs. We measured, through symmetry index, (i) motor parameters: APAs and stepping, and (ii) cortical activity: the PSD of the frontal, sensorimotor and occipital areas during APA, STEP-I (moment of heel-off of the leading foot in the GI until the heel contact of the same foot); and STEP-II (moment of the heel-off of the trailing foot in the GI until the heel contact of the same foot) phases.

Results

Parkinson's disease showed higher asymmetry in cortical activity during APA, STEP-I and STEP-II phases and step velocity (STEP-II phase) during unobstructed GI than CG. However, unexpectedly, PwPD reduced the level of asymmetry of anterior-posterior displacement (p < 0.01) and medial-lateral velocity (p < 0.05) of the APAs. Also, when an obstacle was in place, PwPD showed higher APAs asymmetry (medial-lateral velocity: p < 0.002), with reduced and increased asymmetry of the cortical activity during APA and STEP-I phases, respectively.

Conclusion

Parkinson's disease were not motor asymmetric during GI, indicating that higher cortical activity asymmetry can be interpreted as an adaptive behavior to reduce motor asymmetry. In addition, the presence of obstacle did not regulate motor asymmetry during GI in PwPD.

NA-OR: A path optimization method for manipulators via node attraction and obstacle repulsion

This paper is concerned with the issue of path optimization for manipulators in multi-obstacle environments. Aimed at overcoming the deficiencies of the sampling-based path planning algorithm with high path curvature and low safety margin, a path optimization method, named NA-OR, is proposed for manipulators, where the NA (node attraction) and OR (obstacle repulsion) functions are developed to refine the path by iterations. In the iterations of path optimization, the node attraction function is designed to pull the path nodes toward the center of their neighbor nodes, thereby reducing the path curvature and improving the smoothness. Also, the obstacle repulsion function is developed to push the path nodes out of the potentially unsafe region by generating a repulsive torque on the path nodes, thus improving the safety margin of the motion. By introducing the effect of NA-OR, the optimized path has a significant improvement in path curvature and safety margin compared with the initial path planned by Bi-RRT, which meaningfully enhances the operation ability of manipulators for the applications that give a strong emphasis on security. Experimental results on a 6-DOF manipulator in 4 scenarios demonstrate the effectiveness and superiority of the proposed method in terms of the path cost, safety margin, and path smoothness.

A Study on Dynamic Motion Planning for Autonomous Vehicles Based on Nonlinear Vehicle Model

Autonomous driving technology, especially motion planning and the trajectory tracking method, is the foundation of an intelligent interconnected vehicle, which needs to be improved urgently. Currently, research on path planning methods has improved, but few of the current studies consider the vehicle's nonlinear characteristics in the reference model, due to the heavy computational effort. At present, most of the algorithms are designed by a linear vehicle model in order to achieve the real-time performance at the cost of lost accuracy. To achieve a better performance, the dynamics and kinematics characteristics of the vehicle must be simulated, and real-time computing ensured at the same time. In this article, a Takagi-Sugeno fuzzy-model-based closed-loop rapidly exploring random tree algorithm with on-line re-planning process is applied to build the motion planner, which effectively improves the vehicle performance of dynamic obstacle avoidance, and plans the local obstacle avoidance path in line with the dynamic characteristics of the vehicle. A nonlinear vehicle model is integrated into the motion planner design directly. For fast local path planning mission, the Takagi-Sugeno fuzzy modelling method is applied to the modeling process in the planner design, so that the vehicle state can be directly utilized into the path planner to create a feasible path in real-time. The performance of the planner was evaluated by numerical simulation. The results demonstrate that the proposed motion planner can effectively generate a reference trajectory that guarantees driving efficiency with a lower re-planning rate.

Probing the deployment of peripheral visual attention during obstacle-crossing planning

Gaze is directed to one location at a time, making peripheral visual input important for planning how to negotiate different terrain during walking. Whether and how the brain attends to this input is unclear. We developed a novel paradigm to probe the deployment of sustained covert visual attention by testing orientation discrimination of a Gabor patch at stepping and non-stepping locations during obstacle-crossing planning. Compared to remaining stationary, obstacle-crossing planning decreased visual performance (percent correct) and sensitivity (d') at only the first of two stepping locations. Given the timing of the first and second steps before obstacle crossing relative to the Gabor patch presentation, the results suggest the brain uses peripheral vision to plan one step at a time during obstacle crossing, in contrast to how it uses central vision to plan two or more steps in advance. We propose that this protocol, along with multiple possible variations, presents a novel behavioral approach to identify the role of covert visual attention during obstacle-crossing planning and other goal-directed walking tasks.

A human-in-the-loop approach for enhancing mobile robot navigation in presence of obstacles not detected by the sensory set

Human-in-the-loop approaches can greatly enhance the human-robot interaction by making the user an active part of the control loop, who can provide a feedback to the robot in order to augment its capabilities. Such feedback becomes even more important in all those situations where safety is of utmost concern, such as in assistive robotics. This study aims to realize a human-in-the-loop approach, where the human can provide a feedback to a specific robot, namely, a smart wheelchair, to augment its artificial sensory set, extending and improving its capabilities to detect and avoid obstacles. The feedback is provided by both a keyboard and a brain-computer interface: with this scope, the work has also included a protocol design phase to elicit and evoke human brain event-related potentials. The whole architecture has been validated within a simulated robotic environment, with electroencephalography signals acquired from different test subjects.

Starling-Behavior-Inspired Flocking Control of Fixed-Wing Unmanned Aerial Vehicle Swarm in Complex Environments with Dynamic Obstacles

For the sake of accomplishing the rapidity, safety and consistency of obstacle avoidance for a large-scale unmanned aerial vehicle (UAV) swarm in a dynamic and unknown 3D environment, this paper proposes a flocking control algorithm that mimics the behavior of starlings. By analyzing the orderly and rapid obstacle avoidance behavior of a starling flock, a motion model inspired by a flock of starlings is built, which contains three kinds of motion patterns, including the collective pattern, evasion pattern and local-following pattern. Then, the behavior patterns of the flock of starlings are mapped on a fixed-wing UAV swarm to improve the ability of obstacle avoidance. The key contribution of this paper is collective and collision-free motion planning for UAV swarms in unknown 3D environments with dynamic obstacles. Numerous simulations are conducted in different scenarios and the results demonstrate that the proposed algorithm improves the speed, order and safety of the UAV swarm when avoiding obstacles.

Finger-Gesture Controlled Wheelchair with Enabling IoT

Modern wheelchairs, with advanced and robotic technologies, could not reach the life of millions of disabled people due to their high costs, technical limitations, and safety issues. This paper proposes a gesture-controlled smart wheelchair system with an IoT-enabled fall detection mechanism to overcome these problems. It can recognize gestures using Convolutional Neural Network (CNN) model along with computer vision algorithms and can control the wheelchair automatically by utilizing these gestures. It maintains the safety of the users by performing fall detection with IoT-based emergency messaging systems. The development cost of the overall system is cheap and is lesser than USD 300. Hence, it is expected that the proposed smart wheelchair should be affordable, safe, and helpful to physically disordered people in their independent mobility.

Pathophysiological and motor factors associated with collision avoidance behavior in individuals with stroke

BACKGROUND

High collision rates and frequency of entering the opening from non-paretic sides are associated with collision in individuals with stroke.

OBJECTIVE

To identify factors associated with collision avoidance behavior when individuals with stroke walked through narrow openings.

METHODS

Participants with subacute or chronic stroke walked through a narrow opening and had to avoid colliding with obstacles. Multiple regression analyses were conducted with pathophysiology, motor function, and judgment ability as predictor variables; collision rate and frequency of entering the opening from non-paretic sides were outcome variables.

RESULTS

Sixty-one eligible individuals with stroke aged 63±12 years were enrolled. Thirty participants collided twice or more and 37 entered the opening from the non-paretic side. Higher collision occurrence was associated with slower Timed Up and Go tests and left-right sway (odds ratios, 1.2 and 5.6; 95% confidence intervals, 1.1-1.3 and 1.3-28.2; p = .008 and.025, respectively). Entering from non-paretic sides was associated with lesions in the thalamus, left-sided hemiplegia, and Brunnstrom stage 3 or lower (odds ratios, 6.6, 8.7, and 6.7; 95% confidence intervals, 1.3-52.5, 2.5-36.5, and 1.2-57.5; and p = .038,.001, and.048, respectively).

CONCLUSION

Walking ability is associated with avoiding obstacle collision, while pathophysiological characteristics and degree of paralysis are associated with a preference for which side of the body enters an opening first. Interventions to improve walking ability may improve collision avoidance. Avoidance behavior during intervention varies depending on the lesion position.

Deep Learning-Based NMPC for Local Motion Planning of Last-Mile Delivery Robot

Feasible local motion planning for autonomous mobile robots in dynamic environments requires predicting how the scene evolves. Conventional navigation stakes rely on a local map to represent how a dynamic scene changes over time. However, these navigation stakes depend highly on the accuracy of the environmental map and the number of obstacles. This study uses semantic segmentation-based drivable area estimation as an alternative representation to assist with local motion planning. Notably, a realistic 3D simulator based on an Unreal Engine was created to generate a synthetic dataset under different weather conditions. A transfer learning technique was used to train the encoder-decoder model to segment free space from the occupied sidewalk environment. The local planner uses a nonlinear model predictive control (NMPC) scheme that inputs the estimated drivable space, the state of the robot, and a global plan to produce safe velocity commands that minimize the tracking cost and actuator effort while avoiding collisions with dynamic and static obstacles. The proposed approach achieves zero-shot transfer from a simulation to real-world environments that have never been experienced during training. Several intensive experiments were conducted and compared with the dynamic window approach (DWA) to demonstrate the effectiveness of our system in dynamic sidewalk environments.

Path Planning and Formation Control for UAV-Enabled Mobile Edge Computing Network

Recent developments in unmanned aerial vehicles (UAVs) have led to the introduction of a wide variety of innovative applications, especially in the Mobile Edge Computing (MEC) field. UAV swarms are suggested as a promising solution to cope with the issues that may arise when connecting Internet of Things (IoT) applications to a fog platform. We are interested in a crucial aspect of designing a swarm of UAVs in this work, which is the coordination of swarm agents in complicated and unknown environments. Centralized leader-follower formations are one of the most prevalent architectural designs in the literature. In the event of a failed leader, however, the entire mission is canceled. This paper proposes a framework to enable the use of UAVs under different MEC architectures, overcomes the drawbacks of centralized architectures, and improves their overall performance. The most significant contribution of this research is the combination of distributed formation control, online leader election, and collaborative obstacle avoidance. For the initial phase, the optimal path between departure and arrival points is generated, avoiding obstacles and agent collisions. Next, a quaternion-based sliding mode controller is designed for formation control and trajectory tracking. Moreover, in the event of a failed leader, the leader election phase allows agents to select the most qualified leader for the formation. Multiple possible scenarios simulating real-time applications are used to evaluate the framework. The obtained results demonstrate the capability of UAVs to adapt to different MEC architectures under different constraints. Lastly, a comparison is made with existing structures to demonstrate the effectiveness, safety, and durability of the designed framework.

Design and development of orchard autonomous navigation spray system

Driven by the demand for efficient plant protection in orchards, the autonomous navigation system for orchards is hereby designed and developed in this study. According to the three modules of unmanned system "perception-decision-control," the environment perception and map construction strategy based on 3D lidar is constructed for the complex environment in orchards. At the same time, millimeter-wave radar is further selected for multi-source information fusion for the perception of obstacles. The extraction of orchard navigation lines is achieved by formulating a four-step extraction strategy according to the obtained lidar data. Finally, aiming at the control problem of plant protection machine, the ADRC control strategy is adopted to enhance the noise immunity of the system. Different working conditions are designed in the experimental section for testing the obstacle avoidance performance and navigation accuracy of the autonomous navigation sprayer. The experimental results show that the unmanned vehicle can identify the obstacle quickly and make an emergency stop and find a rather narrow feasible area when a moving person or a different thin column is used as an obstacle. Many experiments have shown a safe distance for obstacle avoidance about 0.5 m, which meets the obstacle avoidance requirements. In the navigation accuracy experiment, the average navigation error in both experiments is within 15 cm, satisfying the requirements for orchard spray operation. A set of spray test experiments are designed in the final experimental part to further verify the feasibility of the system developed by the institute, and the coverage rate of the leaves of the canopy is about 50%.

The Intelligent Path Planning System of Agricultural Robot via Reinforcement Learning

Agricultural robots are one of the important means to promote agricultural modernization and improve agricultural efficiency. With the development of artificial intelligence technology and the maturity of Internet of Things (IoT) technology, people put forward higher requirements for the intelligence of robots. Agricultural robots must have intelligent control functions in agricultural scenarios and be able to autonomously decide paths to complete agricultural tasks. In response to this requirement, this paper proposes a Residual-like Soft Actor Critic (R-SAC) algorithm for agricultural scenarios to realize safe obstacle avoidance and intelligent path planning of robots. In addition, in order to alleviate the time-consuming problem of exploration process of reinforcement learning, this paper proposes an offline expert experience pre-training method, which improves the training efficiency of reinforcement learning. Moreover, this paper optimizes the reward mechanism of the algorithm by using multi-step TD-error, which solves the probable dilemma during training. Experiments verify that our proposed method has stable performance in both static and dynamic obstacle environments, and is superior to other reinforcement learning algorithms. It is a stable and efficient path planning method and has visible application potential in agricultural robots.

Birds both avoid and control collisions by harnessing visually guided force vectoring

Birds frequently manoeuvre around plant clutter in complex-structured habitats. To understand how they rapidly negotiate obstacles while flying between branches, we measured how foraging Pacific parrotlets avoid horizontal strings obstructing their preferred flight path. Informed by visual cues, the birds redirect forces with their legs and wings to manoeuvre around the obstacle and make a controlled collision with the goal perch. The birds accomplish aerodynamic force vectoring by adjusting their body pitch, stroke plane angle and lift-to-drag ratios beat-by-beat, resulting in a range of about 100° relative to the horizontal plane. The key role of drag in force vectoring revises earlier ideas on how the avian stroke plane and body angle correspond to aerodynamic force direction-providing new mechanistic insight into avian manoeuvring-and how the evolution of flight may have relied on harnessing drag.

Walking through a narrow opening improves collision avoidance behavior in a patient with stroke and unilateral spatial neglect: an ABA single-case design

We investigated the effect of a 3-week intervention-wherein a patient with unilateral spatial neglect walks through a narrow opening while entering from the contralesional side-to improve walking ability or ADL. A 66-year-old man was diagnosed with right parietal subcortical hemorrhage. We used an ABA single-case design; period B was set as the intervention. The intervention improved the continuous walking distance and balance ability and decreased the number of collisions when walking through the narrow opening; however, it exerted minimal effect on ADL. Thus, the intervention may effectively improve continuous physical or spatial attention behavior, regardless of ADL improvement.

Computer Vision-Based Path Planning for Robot Arms in Three-Dimensional Workspaces Using Q-Learning and Neural Networks

Computer vision-based path planning can play a crucial role in numerous technologically driven smart applications. Although various path planning methods have been proposed, limitations, such as unreliable three-dimensional (3D) localization of objects in a workspace, time-consuming computational processes, and limited two-dimensional workspaces, remain. Studies to address these problems have achieved some success, but many of these problems persist. Therefore, in this study, which is an extension of our previous paper, a novel path planning approach that combined computer vision, Q-learning, and neural networks was developed to overcome these limitations. The proposed computer vision-neural network algorithm was fed by two images from two views to obtain accurate spatial coordinates of objects in real time. Next, Q-learning was used to determine a sequence of simple actions: up, down, left, right, backward, and forward, from the start point to the target point in a 3D workspace. Finally, a trained neural network was used to determine a sequence of joint angles according to the identified actions. Simulation and experimental test results revealed that the proposed combination of 3D object detection, an agent-environment interaction in the Q-learning phase, and simple joint angle computation by trained neural networks considerably alleviated the limitations of previous studies.

Vision-Based Automated Control of Magnetic Microrobots

Magnetic microrobots are vital tools for targeted therapy, drug delivery, and micromanipulation on cells in the biomedical field. In this paper, we report an automated control and path planning method of magnetic microrobots based on computer vision. Spherical microrobots can be driven in the rotating magnetic field generated by electromagnetic coils. Under microscopic visual navigation, robust target tracking is achieved using PID-based closed-loop control combined with the Kalman filter, and intelligent obstacle avoidance control can be achieved based on the dynamic window algorithm (DWA) implementation strategy. To improve the performance of magnetic microrobots in trajectory tracking and movement in complicated environments, the magnetic microrobot motion in the flow field at different velocities and different distribution obstacles was investigated. The experimental results showed that the vision-based controller had an excellent performance in a complex environment and that magnetic microrobots could be controlled to move to the target position smoothly and accurately. We envision that the proposed method is a promising opportunity for targeted drug delivery in biological research.

A New Multidimensional Repulsive Potential Field to Avoid Obstacles by Nonholonomic UAVs in Dynamic Environments

The ability of autonomous flight with obstacle avoidance should be a fundamental feature of all modern unmanned aerial vehicles (UAVs). The complexity and difficulty of such a task, however, significantly increase in cases combining moving obstacles and nonholonomic UAVs. Additionally, since they assume the symmetrical distribution of repulsive forces around obstacles, traditional repulsive potential fields are not well suited for nonholonomic vehicles. The limited maneuverability of these types of UAVs, including fixed-wing aircraft, requires consideration not only of their relative position, but also their speed as well as the direction in which the obstacles are moving. To address this issue, the following work presents a novel multidimensional repulsive potential field dedicated to nonholonomic UAVs. This field generates forces that repulse the UAV not from the obstacle’s geometrical center, but from areas immediately behind and in front of it located along a line defined by the obstacle’s velocity vector. The strength of the repulsive force depends on the UAV’s distance to the line representing the obstacle’s movement direction, distance to the obstacle along that line, and the relative speed between the UAV and the obstacle projected to the line, making the proposed repulsive potential field multidimensional. Numerical simulations presented within the paper prove the effectiveness of the proposed novel repulsive potential field in controlling the flight of nonholonomic UAVs.

Obstacle Avoidance of Multi-Sensor Intelligent Robot Based on Road Sign Detection

The existing ultrasonic obstacle avoidance robot only uses an ultrasonic sensor in the process of obstacle avoidance, which can only be avoided according to the fixed obstacle avoidance route. Obstacle avoidance cannot follow additional information. At the same time, existing robots rarely involve the obstacle avoidance strategy of avoiding pits. In this study, on the basis of ultrasonic sensor obstacle avoidance, visual information is added so the robot in the process of obstacle avoidance can refer to the direction indicated by road signs to avoid obstacles, at the same time, the study added an infrared ranging sensor, so the robot can avoid potholes. Aiming at this situation, this paper proposes an intelligent obstacle avoidance design of an autonomous mobile robot based on a multi-sensor in a multi-obstruction environment. A CascadeClassifier is used to train positive and negative samples for road signs with similar color and shape. A multi-sensor information fusion is used for path planning and the obstacle avoidance logic of the intelligent robot is designed to realize autonomous obstacle avoidance. The infrared sensor is used to obtain the environmental information of the ground depression on the wheel path, the ultrasonic sensor is used to obtain the distance information of the surrounding obstacles and road signs, and the information of the road signs obtained by the camera is processed by the computer and transmitted to the main controller. The environment information obtained is processed by the microprocessor and the control command is output to the execution unit. The feasibility of the design is verified by analyzing the distance acquired by the ultrasonic sensor, infrared distance measuring sensors, and the model obtained by training the sample of the road sign, as well as by experiments in the complex environment constructed manually.

Navigational Behavior of Humans and Deep Reinforcement Learning Agents

Rapid advances in the field of Deep Reinforcement Learning (DRL) over the past several years have led to artificial agents (AAs) capable of producing behavior that meets or exceeds human-level performance in a wide variety of tasks. However, research on DRL frequently lacks adequate discussion of the low-level dynamics of the behavior itself and instead focuses on meta-level or global-level performance metrics. In doing so, the current literature lacks perspective on the qualitative nature of AA behavior, leaving questions regarding the spatiotemporal patterning of their behavior largely unanswered. The current study explored the degree to which the navigation and route selection trajectories of DRL agents (i.e., AAs trained using DRL) through simple obstacle ridden virtual environments were equivalent (and/or different) from those produced by human agents. The second and related aim was to determine whether a task-dynamical model of human route navigation could not only be used to capture both human and DRL navigational behavior, but also to help identify whether any observed differences in the navigational trajectories of humans and DRL agents were a function of differences in the dynamical environmental couplings.

Obstacle Avoidance Path Planning for Worm-like Robot Using Bézier Curve

Worm-like robots have demonstrated great potential in navigating through environments requiring body shape deformation. Some examples include navigating within a network of pipes, crawling through rubble for search and rescue operations, and medical applications such as endoscopy and colonoscopy. In this work, we developed path planning optimization techniques and obstacle avoidance algorithms for the peristaltic method of locomotion of worm-like robots. Based on our previous path generation study using a modified rapidly exploring random tree (RRT), we have further introduced the Bézier curve to allow more path optimization flexibility. Using Bézier curves, the path planner can explore more areas and gain more flexibility to make the path smoother. We have calculated the obstacle avoidance limitations during turning tests for a six-segment robot with the developed path planning algorithm. Based on the results of our robot simulation, we determined a safe turning clearance distance with a six-body diameter between the robot and the obstacles. When the clearance is less than this value, additional methods such as backward locomotion may need to be applied for paths with high obstacle offset. Furthermore, for a worm-like robot, the paths of subsequent segments will be slightly different than the path of the head segment. Here, we show that as the number of segments increases, the differences between the head path and tail path increase, necessitating greater lateral clearance margins.