Immediate Pose Recovery Method for Untracked Frames in Feature-Based SLAM

In challenging environments, feature-based visual SLAM encounters frequent failures in frame tracking, introducing unknown poses to robotic applications. This paper introduces an immediate approach for recovering untracked camera poses. Through the retrieval of key information from elapsed untracked frames, lost poses are efficiently restored with a short time consumption. Taking account of reconstructed poses and map points during local optimizing, a denser local map is constructed around ambiguous frames to enhance the further SLAM procedure. The proposed method is implemented in a SLAM system, and monocular experiments are conducted on datasets. The experimental results demonstrate that our method can reconstruct the untracked frames in nearly real time, effectively complementing missing segments of the trajectory. Concurrently, the accuracy and robustness for subsequent tracking are improved through the integration of recovered poses and map points.

Heuristics and Learning Models for Dubins MinMax Traveling Salesman Problem

This paper addresses a MinMax variant of the Dubins multiple traveling salesman problem (mTSP). This routing problem arises naturally in mission planning applications involving fixed-wing unmanned vehicles and ground robots. We first formulate the routing problem, referred to as the one-in-a-set Dubins mTSP problem (MD-GmTSP), as a mixed-integer linear program (MILP). We then develop heuristic-based search methods for the MD-GmTSP using tour construction algorithms to generate initial feasible solutions relatively fast and then improve on these solutions using variants of the variable neighborhood search (VNS) metaheuristic. Finally, we also explore a graph neural network to implicitly learn policies for the MD-GmTSP using a learning-based approach; specifically, we employ an S-sample batch reinforcement learning method on a shared graph neural network architecture and distributed policy networks to solve the MD-GMTSP. All the proposed algorithms are implemented on modified TSPLIB instances, and the performance of all the proposed algorithms is corroborated. The results show that learning based approaches work well for smaller sized instances, while the VNS based heuristics find the best solutions for larger instances.

Parallax Inference for Robust Temporal Monocular Depth Estimation in Unstructured Environments

Estimating the distance to objects is crucial for autonomous vehicles, but cost, weight or power constraints sometimes prevent the use of dedicated depth sensors. In this case, the distance has to be estimated from on-board mounted RGB cameras, which is a complex task especially for environments such as natural outdoor landscapes. In this paper, we present a new depth estimation method suitable for use in such landscapes. First, we establish a bijective relationship between depth and the visual parallax of two consecutive frames and show how to exploit it to perform motion-invariant pixel-wise depth estimation. Then, we detail our architecture which is based on a pyramidal convolutional neural network where each level refines an input parallax map estimate by using two customized cost volumes. We use these cost volumes to leverage the visual spatio-temporal constraints imposed by motion and make the network robust for varied scenes. We benchmarked our approach both in test and generalization modes on public datasets featuring synthetic camera trajectories recorded in a wide variety of outdoor scenes. Results show that our network outperforms the state of the art on these datasets, while also performing well on a standard depth estimation benchmark.

Wireless Power Transfer: Systems, Circuits, Standards, and Use Cases

Wireless power transfer provides a most convenient solution to charge devices remotely and without contacts. R&D has advanced the capabilities, variety, and maturity of solutions greatly in recent years. This survey provides a comprehensive overview of the state of the art on different technological concepts, including electromagnetic coupled and uncoupled systems and acoustic technologies. Solutions to transfer mW to MW of power, over distances ranging from millimeters to kilometers, and exploiting wave concepts from kHz to THz, are covered. It is an attractive charging option for many existing applications and moreover opens new opportunities. Various technologies are proposed to provide wireless power to these devices. The main challenges reside in the efficiency and range of the transfer. We highlight innovation in beamforming and UV-assisted approaches. Of particular interest for designers is the discussion of implementation and operational aspects, standards, and safety relating to regulations. A high-level catalog of potential applications maps these to adequate technological options for wireless power transfer.

A Probabilistic Approach to Estimating Allowed SNR Values for Automotive LiDARs in "Smart Cities" under Various External Influences

The paper proposes an approach to assessing the allowed signal-to-noise ratio (SNR) for light detection and ranging (LiDAR) of unmanned autonomous vehicles based on the predetermined probability of false alarms under various intentional and unintentional influencing factors. The focus of this study is on the relevant issue of the safe use of LiDAR data and measurement systems within the “smart city” infrastructure. The research team analyzed and systematized various external impacts on the LiDAR systems, as well as the state-of-the-art approaches to improving their security and resilience. It has been established that the current works on the analysis of external influences on the LiDARs and methods for their mitigation focus mainly on physical (hardware) approaches (proposing most often other types of modulation and optical signal frequencies), and less often software approaches, through the use of additional anomaly detection techniques and data integrity verification systems, as well as improving the efficiency of data filtering in the cloud point. In addition, the sources analyzed in this paper do not offer methodological support for the design of the LiDAR in the very early stages of their creation, taking into account a priori assessment of the allowed SNR threshold and probability of detecting a reflected pulse and the requirements to minimize the probability of “missing” an object when scanning with no a priori assessments of the detection probability characteristics of the LiDAR. The authors propose a synthetic approach as a mathematical tool for designing a resilient LiDAR system. The approach is based on the physics of infrared radiation, the Bayesian theory, and the Neyman–Pearson criterion. It features the use of a predetermined threshold for false alarms, the probability of interference in the analytics, and the characteristics of the LiDAR’s receivers. The result is the analytical solution to the problem of calculating the allowed SNR while stabilizing the level of “false alarms” in terms of background noise caused by a given type of interference. The work presents modelling results for the “false alarm” probability values depending on the selected optimality criterion. The efficiency of the proposed approach has been proven by the simulation results of the received optical power of the LiDAR’s signal based on the calculated SNR threshold and noise values.

Distributed Architecture for Unmanned Vehicle Services

The demand for online services is increasing. Services that would require a long time to understand, use and master are becoming as transparent as possible to the users, that tend to focus only on the final goals. Combined with the advantages of the unmanned vehicles (UV), from the unmanned factor to the reduced size and costs, we found an opportunity to bring to users a wide variety of services supported by UV, through the Internet of Unmanned Vehicles (IoUV). Current solutions were analyzed and we discussed scalability and genericity as the principal concerns. Then, we proposed a solution that combines several services and UVs, available from anywhere at any time, from a cloud platform. The solution considers a cloud distributed architecture, composed by users, services, vehicles and a platform, interconnected through the Internet. Each vehicle provides to the platform an abstract and generic interface for the essential commands. Therefore, this modular design makes easier the creation of new services and the reuse of the different vehicles. To confirm the feasibility of the solution we implemented a prototype considering a cloud-hosted platform and the integration of custom-built small-sized cars, a custom-built quadcopter, and a commercial Vertical Take-Off and Landing (VTOL) aircraft. To validate the prototype and the vehicles' remote control, we created several services accessible via a web browser and controlled through a computer keyboard. We tested the solution in a local network, remote networks and mobile networks (i.e., 3G and Long-Term Evolution (LTE)) and proved the benefits of decentralizing the communications into multiple point-to-point links for the remote control. Consequently, the solution can provide scalable UV-based services, with low technical effort, for anyone at anytime and anywhere.

Influencing Trust for Human-Automation Collaborative Scheduling of Multiple Unmanned Vehicles

OBJECTIVE

We examined the impact of priming on operator trust and system performance when supervising a decentralized network of heterogeneous unmanned vehicles (UVs).

BACKGROUND

Advances in autonomy have enabled a future vision of single-operator control of multiple heterogeneous UVs. Real-time scheduling for multiple UVs in uncertain environments requires the computational ability of optimization algorithms combined with the judgment and adaptability of human supervisors. Because of system and environmental uncertainty, appropriate operator trust will be instrumental to maintain high system performance and prevent cognitive overload.

METHOD

Three groups of operators experienced different levels of trust priming prior to conducting simulated missions in an existing, multiple-UV simulation environment.

RESULTS

Participants who play computer and video games frequently were found to have a higher propensity to overtrust automation. By priming gamers to lower their initial trust to a more appropriate level, system performance was improved by 10% as compared to gamers who were primed to have higher trust in the automation.

CONCLUSION

Priming was successful at adjusting the operator's initial and dynamic trust in the automated scheduling algorithm, which had a substantial impact on system performance.

APPLICATION

These results have important implications for personnel selection and training for futuristic multi-UV systems under human supervision. Although gamers may bring valuable skills, they may also be potentially prone to automation bias. Priming during training and regular priming throughout missions may be one potential method for overcoming this propensity to overtrust automation.