Heuristic Planning for Rough Terrain Locomotion in Presence of External Disturbances and Variable Perception Quality

Abstract

The quality of visual feedback can vary significantly on a legged robot meant to traverse unknown and unstructured terrains. The map of the environment, acquired with online state-of-the-art algorithms, often degrades after a few steps, due to sensing inaccuracies, slippage and unexpected disturbances. If a locomotion algorithm is not designed to deal with this degradation, its planned trajectories might endup to be inconsistent in reality. In this work, we propose a heuristic-based planning approach that enables a quadruped robot to successfully traverse a significantly rough terrain (e.g. stones up to 10 cm of diameter), in absence of visual feedback. When available, the approach allows also to leverage the visual feedback (e.g. to enhance the stepping strategy) in multiple ways, according to the quality of the 3D map. The proposed framework also includes reflexes, triggered in specific situations, and the possibility to estimate online an unknown time-varying disturbance and compensate for it. We demonstrate the effectiveness of the approach with experiments performed on our quadruped robot HyQ (85 kg), traversing different terrains, such as: ramps, rocks, bricks, pallets and stairs. We also demonstrate the capability to estimate and compensate for external disturbances by showing the robot walking up a ramp while pulling a cart attached to its back.

Passive Whole-Body Control for Quadruped Robots: Experimental Validation Over Challenging Terrain

“…The WBC and torque control loops run in real-time threads at 250 Hz and 1 kHz, respectively. We set the gains for the swing tasks to K sw = diag (2000,2000,2000) and D sw = diag (20,20,20), while for the trunk task we set K x = diag (2000,2000,2000) D x = diag(400, 400, 400) and K θ = diag(1000, 1000, 1000) D θ = diag (200,200,200). These values proved to be working in both: simulation and real experiments.…”

“…surface normal). Finally, there is a momentum-based observer that estimates external disturbances [20] and a torque controller.…”

“…l 2 -norms of linear and angular errors are shown in the top and bottom figures. Note that the errors are reset to zero at each step due to re-planning [20]. 15 Setting the weighting matrix R kk = JstS T Rτ SJ T st where: R kk is the sub-block of R to the GRFs variables and S selects the actuation joints.…”

“…These experiments have been performed with online terrain mapping [20]. Both the terrain mapping and the whole-body controller make use of a drift-free state estimation algorithm to obtain the body state [21].…”

STANCE: Locomotion Adaptation Over Soft Terrain

“…Despite the drifting problem, we used the current state estimator [32] in our WBC because the planner in Fig. 3 has a re-planning feature that makes our WBC robust against a drifting state estimator [33]. However, the TCE still requires an accurate and non drifting estimate of the feet position in the world frame.…”