Perceptive whole‐body planning for multilegged robots in confined spaces

Abstract

Legged robots are exceedingly versatile and have the potential to navigate complex, confined spaces due to their many degrees of freedom. As a result of the computational complexity, there exist no online planners for perceptive whole‐body locomotion of robots in tight spaces. In this paper, we present a new method for perceptive planning for multilegged robots, which generates body poses, footholds, and swing trajectories for collision avoidance. Measurements from an onboard depth camera are used to create a three‐dimensional map of the terrain around the robot. We randomly sample body poses then smooth the resulting trajectory while satisfying several constraints, such as robot kinematics and collision avoidance. Footholds and swing trajectories are computed based on the terrain, and the robot body pose is optimized to ensure stable locomotion while not colliding with the environment. Our method is designed to run online on a real robot and generate trajectories several meters long. We first tested our algorithm in several simulations with varied confined spaces using the quadrupedal robot ANYmal. We also simulated experiments with the hexapod robot Weaver to demonstrate applicability to different legged robot configurations. Then, we demonstrated our whole‐body planner in several online experiments both indoors and in realistic scenarios at an emergency rescue training facility. ANYmal, which has a nominal standing height of 80 cm and a width of 59 cm, navigated through several representative disaster areas with openings as small as 60 cm. Three‐meter trajectories were replanned with 500 ms update times.

Whole-Body MPC and Online Gait Sequence Generation for Wheeled-Legged Robots

“…The coordination of each task's solution is one of the main challenges, and heuristics are needed to align the foot and torso motions. Many approaches were developed over the last years exploiting these task synergies [5], [6], [8]- [10], [19]- [25].…”

Collision-Free MPC for Legged Robots in Static and Dynamic Scenes

“…1) Environment representation: The environment's representation is a crucial element, and multiple volumetric map representations are available, e.g., octree [3], height-map [4], and euclidean signed distance field (ESDF). ESDFs seem to become increasingly crucial in legged robotics as they can deal with the sensor's noise in an inherent fashion [5], [6]. Directly storing the distance measurements in the representation enables averaging multiple measurements, thereby filtering the sensor's noise.…”

“…In [5], the CHOMP algorithm generates the shortest collision-free trajectory of the foot's swing trajectory, and the work in [6] extends the work to collision-free planning of the torso trajectory. The authors generate a SDF through a heightmap, and overhanging obstacles are handled through a second 2.5D map of the ceiling.…”

“…1) Overhanging obstacles: The robot automatically ducks under overhanging obstacles and returns to its nominal posture once the obstacle is passed. In contrast to the 2.5D height map in [6], our environment representation based on the SDF can handle the full 3D space, including overhanging obstacles.…”

Modelo Cinemático para Atuação Completa de Robôs Hexápodes

“…As pernas na fase de suporte (support phase -contato com a superfície) formam o sistema fechado e as pernas na fase de transferência (transfer phase -balanço) representam o sistema aberto. Usar o corpo além das pernas para aumentar a mobilidade tem sido um desafio, uma limitação significativa em ambientes altamente desestruturados (Buchanan et al 2020).…”