Whole-body hierarchical motion and force control for humanoid robots

Abstract

International audienceRobots acting in human environments usually need to perform multiple motion and force tasks while respecting a set of constraints. When a physical contact with the environment is established, the newly activated force task or contact constraint may interfere with other tasks. The objective of this paper is to provide a control framework that can achieve real-time control of humanoid robots performing both strict and non strict prioritized motion and force tasks. It is a torque-based quasi-static control framework, which handles a dynamically changing task hierarchy with simultaneous priority transitions as well as activation or deactivation of tasks. A quadratic programming problem is solved to maintain desired task hierarchies, subject to constraints. A generalized projector is used to quantitatively regulate how much a task can influence or be influenced by other tasks through the modulation of a priority matrix. By the smooth variations of the priority matrix, sudden hierarchy rearrangements can be avoided to reduce the risk of instability. The effectiveness of this approach is demonstrated on both a simulated and a real humanoid robot

A Whole-Body Control Framework Based on the Operational Space Formulation Under Inequality Constraints via Task-Oriented Optimization

“…Meanwhile, similar to our method, few studies are utilizing an analytically achieved null-space projection matrix for hierarchical task control. QP optimisation is incorporated with analytic null-space projection matrices in [30], however, dynamic consistency cannot be exploited due to the absence of the inertia weighting matrix in the computation of the null-space projection. Recently, the method in [30] is further extended with consideration of dynamic consistency [31], however, it is not straightforward to apply to the underactuated robot with the floating-base, e.g., humanoids.…”

“…QP optimisation is incorporated with analytic null-space projection matrices in [30], however, dynamic consistency cannot be exploited due to the absence of the inertia weighting matrix in the computation of the null-space projection. Recently, the method in [30] is further extended with consideration of dynamic consistency [31], however, it is not straightforward to apply to the underactuated robot with the floating-base, e.g., humanoids. Interestingly, the authors in [32] suggest computing acceleration in the configuration space based on the desired acceleration commands instead of using operational-space dynamics calculation.…”

Modification of gesture-determined-dynamic function with consideration of margins for motion planning of humanoid robots

“…Hong et al who focused on real-time pattern generation utilized the linear pendulum model to solve the walking control problem of MAHRU-R robot with feed-forward and feed-back controller [49]. The desired task hierarchy which is based on quadratic programming was solve by Liu et al to reduce the risk of instability when humanoid robots conducting operations [50]. According to HRP-2 robot, a novel plan for foot placements was proposed by Kanoun et al to formulate the deformation problem as the inverse kinematic problem and to further be described as a locomotion phase of the desired tasks.…”

The CoDyCo Project Achievements and Beyond: Toward Human Aware Whole-Body Controllers for Physical Human Robot Interaction

“…The end result is typically unstable or undesirable whole-body behaviours, and these tasks can be qualified as incompatible. Prioritisation techniques use weighted sums [14], [15], hierarchies [16], [17] or a mix of both [18], [19] to manage task incompatibilities at the whole-body control level, but are difficult to tune and may not actually solve the problem. Given that it is the task reference values which generate the incompatible control objectives, an alternative to prioritisation tuning is to modify the task trajectories and make them compatible as initially suggested in [20].…”