12 April 2023 in IEEE Transactions on Robotics
Optimal control (OC) using inverse dynamics provides numerical benefits, such as coarse optimization, cheaper computation of derivatives, and a high convergence rate. However, to take advantage of these benefits in model predictive control (MPC) for legged robots, it is crucial to handle efficiently its large number of equality constraints. To accomplish this, we first propose a novel approach to handle equality constraints based on nullspace parameterization. Our approach balances optimality, and both dynamics and equality-constraint feasibility appropriately, which increases the basin of attraction to high-quality local minima. To do so, we modify our feasibility-driven search by incorporating a merit function. Furthermore, we introduce a condensed formulation of inverse dynamics that considers arbitrary actuator models. We also propose a novel MPC based on inverse dynamics within a perceptive locomotion framework. Finally, we present a theoretical comparison of OC with forward and inverse dynamics and evaluate both numerically. Our approach enables the first application of inverse-dynamics MPC on hardware, resulting in the state-of-the-art dynamic climbing on the ANYmal robot. We benchmark it over a wide range of robotics problems and generate agile and complex maneuvers. We show the computational reduction of our nullspace resolution and condensed formulation (up to ${47.3}\boldsymbol{\%}$ ). We provide evidence of the benefits of our approach by solving coarse optimization problems with a high convergence rate (up to 10 Hz of discretization). Our algorithm is publicly available inside Crocoddyl .
4 April 2025 in IEEE Access
Real-time synthesis of legged locomotion maneuvers in challenging industrial settings is still an open problem, requiring simultaneous determination of footsteps locations several steps ahead while generating whole-body motions close to the robot’s limits. State estimation and perception errors impose the practical constraint of fast re-planning motions in a model predictive control (MPC) framework. We first observe that the computational limitation of perceptive locomotion pipelines lies in the combinatorics of contact surface selection. Re-planning contact locations on selected surfaces can be accomplished at MPC frequencies (50-100 Hz). Then, whole-body motion generation typically follows a reference trajectory for the robot base to facilitate convergence. We propose removing this constraint to robustly address unforeseen events such as contact slipping, by leveraging a state-of-the-art whole-body MPC (Croccodyl). Our contributions are integrated into a complete framework for perceptive locomotion, validated under diverse terrain conditions, and demonstrated in challenging trials that push the robot’s actuation limits, as well as in the ICRA 2023 quadruped challenge simulation.
19 December 2023 in IEEE Access
Full-dynamics model predictive control (MPC) has recently been applied to quadrupedal locomotion in semi-unstructured environments. These advances have been fueled by the availability of efficient trajectory optimization (TO) algorithms and inexpensive computational power. The main advantages of full-dynamics MPC are (i) enabling complex locomotion manoeuvres, (ii) considering actuation limits, and (iii) improving robot stability. However, to make the TO problem sufficiently simple to be solved at run time, reference swing foot trajectories are usually tracked in the MPC formulation. These trajectories are often computed independently of the motion of the joints, limiting the approach generality and capability. To address this limitation, we present a full-dynamics MPC formulation that does not require reference swing-foot trajectories, featuring a novel cost function targeting swing foot motion and considering environmental information. Removing the need for reference swing foot trajectories, our approach can also automatically adjust footstep locations, as long as the contact surfaces are predefined. We have validated our MPC formulation through simulations and experiments on the ANYmal B robot. Our approach has similar computational efficiency to state-of-the-art formulations, while displaying superior push-recovery capabilities on various terrains.