MPC-based Controller with Terrain Insight for Dynamic Legged Locomotion

TL;DR

This paper introduces a control strategy for dynamic legged robots that integrates terrain insight via neural networks and model predictive control, enabling safe and efficient traversal of complex terrains using onboard sensing.

Contribution

It presents a novel combination of a neural network-based contact sequence task with MPC for terrain-aware legged locomotion, suitable for onboard computation.

Findings

Successful simulation of HyQReal traversing rough terrain

Real-time terrain evaluation with neural networks

Enhanced safety and efficiency in locomotion

Abstract

We present a novel control strategy for dynamic legged locomotion in complex scenarios, that considers information about the morphology of the terrain in contexts when only on-board mapping and computation are available. The strategy is built on top of two main elements: first a contact sequence task that provides safe foothold locations based on a convolutional neural network to perform fast and continuous evaluation of the terrain in search of safe foothold locations; then a model predictive controller that considers the foothold locations given by the contact sequence task to optimize target ground reaction forces. We assess the performance of our strategy through simulations of the hydraulically actuated quadruped robot HyQReal traversing rough terrain under realistic on-board sensing and computing conditions.