Agile Legged-Wheeled Reconfigurable Navigation Planner Applied on the CENTAURO Robot

Abstract

Hybrid legged-wheeled robots such as the CEN-TAURO, are capable of varying their footprint polygon to carry out various agile motions. This property can be advantageous for wheeled-only planning in cluttered spaces, which is our focus. In this paper, we present an improved algorithm that builds upon our previously introduced preliminary footprint varying A* planner, which was based on the rectangular symmetry of the foot support polygon. In particular, we introduce a Theta* based planner with trapezium-like search, which aims to further reduce the limitations imposed upon the wheeled-only navigation of the CENTAURO robot by the low-dimensional search space, maintaining the real-time computational efficiency. The method is tested on the simulated and real full-size CENTAURO robot in cluttered environments.

Task-Consistent Path Planning for Mobile 3D Printing

“…The method proposed makes it likely but does not guarantee the existence of a continuous joint-space trajectory of the manipulator. This could be addressed by integrating this work with Cartesian path planner like Descartes [35] or other advanced path planners [36], [37]. Furthermore, there are examples in the literature that show that IRM computation could be made faster [33] or only light-weight features of IRM used instead [38].…”

Reconfigurable and Agile Legged-Wheeled Robot Navigation in Cluttered Environments With Movable Obstacles

“…A further check is done to see if the motion from the parent to child node can be done omni-directionally. Detailed explanations of the variations of the height calculations and the cost functions compared to Prototype 1 can be found in [6].…”

“…This is done by utilizing costmaps, Octomap filtering [29], and simple 2D obstacle segmentation, in order to find safe robot polygons. The search space is then reduced either using rectangular footprint polygonbased search in an A* framework [5] or a more flexible trapezium-like footprint search in the more versatile Theta* planner framework [6] to allow any-angle motion as well as independent motion for the front and back wheel pairs. We use Lidar point cloud data in order to create costmaps and segmentations in the 2D projections of the 3D point clouds.…”

“…We begin, by first determining the push contact surfaces, using the edge pixels coordinates of the object. Since the resolution of the image is 5cm per pixel and the length of the face of the CENTAURO foot is 25cm (justification of the selected values in the global planner work [6]), we search all the edge pixels of the obstacle in the image in sets of 5 in order to find smooth surfaces. The angle between vectors from first to the third pixel and from the fifth to the third pixel is used to determine if the surface is smooth or not.…”

“…This paper, is split into two parts. Firstly, we describe our previously introduced work on a novel global path planner for hybrid wheeled-legged navigation [5], [6]. The goal of the method is to reconfigure the footprint polygon of the robot, by tweaking the leg configuration and the height of the base from the ground, in order to navigate through obstacles that cannot be avoided.…”