Autotuning Bipedal Locomotion MPC with GRFM-Net for Efficient Sim-to-Real Transfer
This work addresses the sim-to-real transfer challenge for humanoid robot locomotion, offering an incremental improvement in parameter optimization efficiency.
The paper tackled the problem of labor-intensive manual tuning in bipedal locomotion control for humanoid robots by developing an autotuning method using DiffTune with GRFM-Net, resulting in a 40.5% reduction in total loss compared to expert-tuned parameters in hardware experiments.
Bipedal locomotion control is essential for humanoid robots to navigate complex, human-centric environments. While optimization-based control designs are popular for integrating sophisticated models of humanoid robots, they often require labor-intensive manual tuning. In this work, we address the challenges of parameter selection in bipedal locomotion control using DiffTune, a model-based autotuning method that leverages differential programming for efficient parameter learning. A major difficulty lies in balancing model fidelity with differentiability. We address this difficulty using a low-fidelity model for differentiability, enhanced by a Ground Reaction Force-and-Moment Network (GRFM-Net) to capture discrepancies between MPC commands and actual control effects. We validate the parameters learned by DiffTune with GRFM-Net in hardware experiments, which demonstrates the parameters' optimality in a multi-objective setting compared with baseline parameters, reducing the total loss by up to 40.5$\%$ compared with the expert-tuned parameters. The results confirm the GRFM-Net's effectiveness in mitigating the sim-to-real gap, improving the transferability of simulation-learned parameters to real hardware.