Reactive Navigation on Natural Environments by Continuous Classification of Ground Traversability

Reactivity is a key component for autonomous vehicles navigating on natural terrains in order to safely avoid unknown obstacles. To this end, it is necessary to continuously assess traversability by processing on-board sensor data. This paper describes the case study of mobile robot Andabata that classifies traversable points from 3D laser scans acquired in motion of its vicinity to build 2D local traversability maps. Realistic robotic simulations with Gazebo were employed to appropriately adjust reactive behaviors. As a result, successful navigation tests with Andabata using the robot operating system (ROS) were performed on natural environments at low speeds.

Traversability Assessment and Trajectory Planning of Unmanned Ground Vehicles with Suspension Systems on Rough Terrain

This paper presents a traversability assessment method and a trajectory planning method. They are key features for the navigation of an unmanned ground vehicle (UGV) in a non-planar environment. In this work, a 3D light detection and ranging (LiDAR) sensor is used to obtain the geometric information about a rough terrain surface. For a given SE(2) pose of the vehicle and a specific vehicle model, the SE(3) pose of the vehicle is estimated based on LiDAR points, and then a traversability is computed. The traversability tells the vehicle the effects of its interaction with the rough terrain. Note that the traversability is computed on demand during trajectory planning, so there is not any explicit terrain discretization. The proposed trajectory planner finds an initial path through the non-holonomic A*, which is a modified form of the conventional A* planner. A path is a sequence of poses without timestamps. Then, the initial path is optimized in terms of the traversability, using the method of Lagrange multipliers. The optimization accounts for the model of the vehicle's suspension system. Therefore, the optimized trajectory is dynamically feasible, and the trajectory tracking error is small. The proposed methods were tested in both the simulation and the real-world experiments. The simulation experiments were conducted in a simulator called Gazebo, which uses a physics engine to compute the vehicle motion. The experiments were conducted in various non-planar experiments. The results indicate that the proposed methods could accurately estimate the SE(3) pose of the vehicle. Besides, the trajectory cost of the proposed planner was lower than the trajectory costs of other state-of-the-art trajectory planners.

Variational Hilbert regression for terrain modeling and trajectory optimization

The ability to generate accurate terrain models is of key importance in a wide variety of robotics tasks, ranging from path planning and trajectory optimization to environment exploration and mining applications. This paper introduces a novel regression methodology for terrain modeling that can approximate arbitrarily complex functions based on a series of simple kernel calculations, using variational Bayesian inference. A sparse feature vector is used to efficiently project input points into a high-dimensional reproducing kernel Hilbert space, according to a set of inducing points automatically generated from clustering available data. Each inducing point maintains its own regression model in addition to individual kernel parameters, and the entire set is iteratively optimized as more data are collected in order to maximize a global variational lower bound. We also show how kernel and regression model parameters can be jointly learned, to achieve a better approximation of the underlying function. Experimental results show that the proposed methodology consistently outperforms current state-of-the-art techniques, while producing a continuous model with a fully probabilistic treatment of uncertainties, well-defined gradients, and highly scalable to large-scale datasets. As a practical application of the proposed terrain modeling technique, we explore the problem of trajectory optimization, deriving gradients that allow the efficient generation of continuous paths using standard optimization algorithms, minimizing a series of useful properties (i.e. distance traveled, changes in elevation, and terrain variance).

3D Position Tracking in Challenging Terrain

The intent of this paper is to demonstrate how the accuracy of 3D position tracking can be improved by considering rover locomotion in rough terrain as a holistic problem. Although the selection of good sensors is crucial to accurately track the rover's position, it is not the only aspect to consider. Indeed, the use of an unadapted locomotion concept severely affects the signal to noise ratio of the sensors, which leads to poor motion estimates. In this work, a mechanical structure allowing smooth motion across obstacles with limited wheel slip is used. In particular, this enables the use of odometry and inertial sensors to improve the position estimation in rough terrain. A method for computing 3D motion increments based on the wheel encoders and chassis state sensors is developed. Because it accounts for the kinematics of the rover, this method provides better results than the standard approach. To further improve the accuracy of the position tracking and the rover's climbing performance, a controller minimizing wheel slip has been developed. The algorithm runs online and can be adapted to any kind of passive wheeled rover. Finally, sensor fusion using 3D-Odometry, inertial sensors and visual motion estimation based on stereovision is presented. The experimental results demonstrate how each sensor contributes to increase the accuracy and robustness of the 3D position estimation.

DenseSLAM: Simultaneous Localization and Dense Mapping

This paper addresses the problem of environment representation for Simultaneous Localization and Mapping (SLAM) algorithms.

One of the main problems of SLAM is how to interpret and synthesize the external sensory information into a representation of the environment that can be used by the mobile robot to operate autonomously. Traditionally, SLAM algorithms have relied on sparse environment representations. However, for autonomous navigation, a more detailed representation of the environment is necessary, and the classic feature-based representation fails to provide a robot with sufficient information. While a dense representation is desirable, it has not been possible for SLAM paradigms.

This paper presents DenseSLAM, an algorithm to obtain and maintain detailed environment representations. The algorithm represents different sensory information in dense multi-layered maps. Each layer can represent different properties of the environment, such as occupancy, traversability, elevation or each layer can describe the same environment property using different representations. Implementations of the algorithm with two different representations for the dense maps are shown.

A rich representation has several potential advantages to assist the navigation process, for example to facilitate data association using multi-dimensional maps. This paper presents two particular applications to improve the localization process; the extraction of complex landmarks from the dense maps and the detection of areas with dynamic objects. The paper also presents an analysis of consistency of the maps obtained with DenseSLAM. The position error in the dense maps is analyzed and a method to select the landmarks in order to minimize these errors is explained.

The algorithm was tested with outdoor experimental data taken with a ground vehicle. The experimental results show that the algorithm can obtain dense environment representations and that the detailed representation can be used to improve the vehicle localization process.

An Efficient Extension to Elevation Maps for Outdoor Terrain Mapping and Loop Closing

Elevation maps are a popular data structure for representing the environment of a mobile robot operating outdoors or on not-flat surfaces. Elevation maps store in each cell of a discrete grid the height of the surface at the corresponding place in the environment. However, the use of this 2½-dimensional representation, is disadvantageous when utilized for mapping with mobile robots operating on the ground, since vertical or overhanging objects cannot be represented appropriately. Furthermore, such objects can lead to registration errors when two elevation maps have to be matched. In this paper, an approach is proposed that allows a mobile robot to deal with vertical and overhanging objects in elevation maps. The approach classifies the points in the environment according to whether they correspond to such objects or not. Also presented is a variant of the ICP algorithm that utilizes the classification of cells during the data association. Additionally, it is shown how the constraints computed by the ICP algorithm can be applied to determine globally consistent alignments. Experiments carried out with a real robot in an outdoor environment demonstrate that the proposed approach yields highly accurate elevation maps even in the case of loops. Experimental results are presented demonstrating that that the proposed classification increases the robustness of the scan matching process.

Extending the Path-Planning Horizon

The mobility sensors on a typical mobile robot vehicle have limited range. Therefore a navigation system has no knowledge about the world beyond this sensing horizon. As a result, path planners that rely only on this knowledge to compute paths are unable to anticipate obstacles sufficiently early and have no choice but to resort to an inefficient local obstacle avoidance behavior.

To alleviate this problem, we present an opportunistic navigation and view planning strategy that incorporates look-ahead sensing of possible obstacle configurations. This planning strategy is based on a “what-if” analysis of hypothetical future configurations of the environment. Candidate sensing positions are evaluated based on their ability to observe anticipated obstacles. These sensing positions identified by this forward-simulation framework are used by the planner as intermediate waypoints. The validity of the strategy is supported by results from simulations as well as field experiments with a real robotic platform. These results show that significant reduction in path length can be achieved by using this framework.

Agile Wheeled Mobile Robots for Service in Natural Environment

Although the wheeled locomotion proved to be very efficient on smooth grounds, it still encounters great difficulties in natural environments, where the ground is subject to wide variations in term of geometry (irregular surface, presence of obstacles...) and material properties (cohesion, grip condition...). This chapter presents recent developments and original systems that improve the capacities of wheeled mobile service robots on natural ground. First is considered the case of low speed motion. Section 2 presents recent results on reconfigurable suspensions that have two states and can decrease lateral friction and energy consumption during turns for skid-steering vehicles. Section 3 presents an original hybrid kinematics that combines wheels with an articulated frame for creating a mobile-wheeled robot with high obstacle-climbing capacities, using only one supplemental actuator. Other advances deal with high-speed motion. Section 4 describes a new device dedicated to vehicle dynamic stability, which improves lateral stability on fast mobile robots during turns and contributes to rollover prevention. Finally, Section 5 introduces innovative suspensions with two DOF for fast obstacle crossing. They damp vertical shocks, such as ordinary suspensions, but also horizontal ones, contributing to tip-over prevention on irregular grounds that feature many steep obstacles.

Non-parametric Learning to Aid Path Planning over Slopes

In this paper we address the problem of closing the loop from perception to action selection for unmanned ground vehicles, with a focus on navigating slopes. A new non-parametric learning technique is presented to generate a mobility representation where the maximum feasible speed is used as a criterion to classify the world. The inputs to the algorithm are terrain gradients derived from an elevation map and past observations of wheel slip. It is argued that such a representation can aid in path planning with improved selection of vehicle heading and velocity in off-road slopes. In addition, an information theoretic test is proposed to validate a chosen proprioceptive representation (such as slip) for mobility map generation. Results of mobility map generation and its benefits to path planning are shown.

Navigating a mobile robot by a traversability field histogram

This paper presents an autonomous terrain navigation system for a mobile robot. The system employs a two-dimensional laser range finder (LRF) for terrain mapping. A so-called "traversability field histogram" (TFH) method is proposed to guide the robot. The TFH method first transforms a local terrain map surrounding the robot's momentary position into a traversability map by extracting the slope and roughness of a terrain patch through least-squares plane fitting. It then computes a so-called "polar traversability index" (PTI) that represents the overall difficulty of traveling along the corresponding direction. The PTIs are represented in a form of histogram. Based on this histogram, the velocity and steering commands of the robot are determined. The concept of a virtual valley and an exit condition are proposed and used to direct the robot such that it can reach the target with a finite-length path. The algorithm is verified by simulation and experimental results.