Robot Navigation In Unstructured Environments Research Articles

Continuous Online Semantic Implicit Representation for Autonomous Ground Robot Navigation in Unstructured Environments

While mobile ground robots have now the physical capacity of travelling in unstructured challenging environments such as extraterrestrial surfaces or devastated terrains, their safe and efficient autonomous navigation has yet to be improved before entrusting them with complex unsupervised missions in such conditions. Recent advances in machine learning applied to semantic scene understanding and environment representations, coupled with modern embedded computational means and sensors hold promising potential in this matter. This paper therefore introduces the combination of semantic understanding, continuous implicit environment representation and smooth informed path-planning in a new method named COSMAu-Nav. It is specifically dedicated to autonomous ground robot navigation in unstructured environments and adaptable for embedded, real-time usage without requiring any form of telecommunication. Data clustering and Gaussian processes are employed to perform online regression of the environment topography, occupancy and terrain traversability from 3D semantic point clouds while providing an uncertainty modeling. The continuous and differentiable properties of Gaussian processes allow gradient based optimisation to be used for smooth local path-planning with respect to the terrain properties. The proposed pipeline has been evaluated and compared with two reference 3D semantic mapping methods in terms of quality of representation under localisation and semantic segmentation uncertainty using a Gazebo simulation, derived from the 3DRMS dataset. Its computational requirements have been evaluated using the Rellis-3D real world dataset. It has been implemented on a real ground robot and successfully employed for its autonomous navigation in a previously unknown outdoor environment.

A hybrid obstacle avoidance method for mobile robot navigation in unstructured environment

PurposeWheeled mobile robots (WMR) are the most widely used robots. Avoiding obstacles in unstructured environments, especially dynamic obstacles such as pedestrians, is a serious challenge for WMR. This paper aims to present a hybrid obstacle avoidance method that combines an informed-rapidly exploring random tree* algorithm with a three-dimensional (3D)-object detection approach and model prediction controller (MPC) to conduct obstacle perception, collision-free path planning and obstacle avoidance for WMR in unstructured environments.Design/methodology/approachGiven a reference orientation and speed, the hybrid method uses parametric ellipses to represent obstacle expansion boundaries based on the 3D target detection results, and a collision-free reference path is planned. Then, the authors build on a model predictive control for tracking the collision-free reference path by incorporating the distance between the robot and obstacles. The proposed framework is a mapless method for WMR.FindingsThe authors present experimental results with a mobile robot for obstacle avoidance in indoor environments crowded with obstacles, such as chairs and pedestrians. The results show that the proposed hybrid obstacle avoidance method can satisfy the application requirements of mobile robots in unstructured environments.Originality/valueIn this study, the parameter ellipse is used to represent the area occupied by the obstacle, which takes the velocity as the parameter. Therefore, the motion direction and position of dynamic obstacles can be considered in the planning stage, which enhances the success rate of obstacle avoidance. In addition, the distance between the obstacle and robot is increased in the MPC optimization function to ensure a safe distance between the robot and the obstacle.

Learning terrain segmentation with classifier ensembles for autonomous robot navigation in unstructured environments

AbstractAutonomous robot navigation in unstructured outdoor environments is a challenging area of active research and is currently unsolved. The navigation task requires identifying safe, traversable paths that allow the robot to progress toward a goal while avoiding obstacles. Stereo is an effective tool in the near field, but used alone leads to a common failure mode in autonomous navigation in which suboptimal trajectories are followed due tonearsightedness, or the robot's inability to distinguish obstacles and safe terrain in the far field. This can be addressed through the use of machine learning methods to accomplishnear‐to‐farlearning, in which near‐field terrain appearance and stereo readings are used to train models able to predict far‐field terrain. This paper proposes to enhance existing,memorylessnear‐to‐far learning approaches through the use ofclassifier ensemblesthat allow terrain models trained on data seen at different points in time to be preserved and referenced later. These stored models serve as memory, and we show that they can be leveraged for more effective far‐field terrain classification on future images seen by the robot. A five‐factor, full‐factorial, repeated‐measures experimental evaluation is performed on hand‐labeled data sets taken directly from the problem domain. The experiments result in many statistically significant findings, the most important being that the proposed near‐to‐far Best‐KEnsemble Algorithm, with appropriate parameter selection, outperforms the single‐model, nonensemble baseline approach in far‐field terrain classification. Several other findings that inform the use of near‐to‐far ensemble methods are also presented. © 2009 Wiley Periodicals, Inc.

Path planning in image space for autonomous robot navigation in unstructured environments

AbstractPath planning systems using graph‐search algorithms such as A* usually operate in uniform plan‐view occupancy grids. However, the sensors used to construct these grids observe the environment in their own sample space based on sensor type and viewpoint. In this paper we present animage spacetechnique for path planning in unknown unstructured outdoor environments. Our method differs from previous techniques in that we perform path search directly in image space—the native sensor space of the imaging sensor. After an image space path has been found, it is used for navigation in the real world. By operating at the resolution of the image sensor, image space planning facilitates accurate robot vs. obstacle localization and enables a high degree of movement precision. Our image space planning techniques can potentially be used with many different kinds of sensor data, and we experimentally evaluate the use of stereo disparity and color information. We present an extension to the basic image space planning system called thecylindrical plannerthat simulates a 2π field of view with a cylindrically shaped occupancy grid. We believe that image space planning is well suited for use in the local subsystem of a hierarchical planner and implement ahybrid hierarchical plannerthat utilizes the cylindrical planner as a local planning subsystem and a two‐dimensional Cartesian planner as the global planning subsystem. All three systems are implemented and experimentally tested on a real robot. We evaluate the failure modes of image space planning and discuss how to avoid them. We find that image space enables precise real‐time, near‐field planning. © 2009 Wiley Periodicals, Inc.

SIMULTANEOUS LOCALIZATION AND MAP BUILDING ALGORITHM FOR REAL-TIME APPLICATIONS

Abstract Mobile robot navigation in unstructured environment is a challenging task due to the uncertain nature of the real world. Navigating using visual landmarks could be a mandatory skill together with the ability of building a representation of the world around the robot. This mapping aptitude should be implemented as an efficient real-time task, even if a large number of elements have to be included in the map itself. To this aim, and to help in localising the robot, a promising technique is given by the Extended Kalman Filter in its interlaced version. The resulting SLAM algorithm, proposed in this paper, has a reduced computational cost preserving, at the same time, a good performance.

Feature extraction and selection for mobile robot navigation in unstructured environments

Feature extraction and selection are important steps in the construction of a map to support the navigation of mobile robots in outdoor environments. The large amount of data acquired by the on-board sensors has to be reduced to retain only the crucial information for the navigation purpose. This procedure should be robust given the rough, dynamic and unpredictable conditions provided by outdoor scenarios. The paper discusses the use of different types of features and proposes a feature selection criteria aiming at building a topological map of the environment. The approach was tested with real data acquired in different types of outdoor scenarios.