When to make a step? Tackling the timing problem in multi-contact locomotion by TOPP-MPC

Stéphane Caron (1) , Quang-Cuong Pham (2)

1 IDH - Interactive Digital Humans
2 Nanyang Technological University [Singapour]

We present a Model-Predictive Controller (MPC) for multi-contact locomotion where predictive optimizations are realized by Time-Optimal Path Parameterization (TOPP). The key feature of this design is that, contrary to existing planners where step timings are provided as inputs, here the timing between contact switches is computed as output to a linear optimization problem based on a dynamic model of the robot. This is particularly appealing to multi-contact locomotion, where proper timings depend on the terrain topology and suitable heuristics are unknown. Thanks to recent advances in multi-contact stability computations, we improve the performance of TOPP for COM trajectories, which allows us to integrate it into a fast control loop. We implement the complete control pipeline and showcase it in simulations where a model of the HRP-4 humanoid climbs up and down a series of hills.

Mots clés: Multi-contact locomotion; Model predictive control; Time-optimal control; Humanoid and Bipedal Locomotion

Domaines: Robotique [cs.RO]

Stéphane Caron, Quang-Cuong Pham. When to make a step? Tackling the timing problem in multi-contact locomotion by TOPP-MPC. 2016. ⟨hal-01363757v1⟩