Receding-Horizon Perceptive Trajectory Optimization for Dynamic Legged Locomotion with Learned Initialization

TL;DR

This paper introduces a receding-horizon trajectory optimization framework for quadruped robots that combines heuristic and learned initializations to enable adaptive, high-speed locomotion over challenging terrain using onboard sensing.

Contribution

It presents a novel pipeline integrating learned trajectory initialization with online optimization for dynamic legged locomotion, improving speed and adaptability.

Findings

Successful obstacle crossing at moderate speeds with onboard sensing.

High-speed locomotion on flat ground demonstrated on real robot.

Reduced optimization effort using learned trajectory regression.

Abstract

To dynamically traverse challenging terrain, legged robots need to continually perceive and reason about upcoming features, adjust the locations and timings of future footfalls and leverage momentum strategically. We present a pipeline that enables flexibly-parametrized trajectories for perceptive and dynamic quadruped locomotion to be optimized in an online, receding-horizon manner. The initial guess passed to the optimizer affects the computation needed to achieve convergence and the quality of the solution. We consider two methods for generating good guesses. The first is a heuristic initializer which provides a simple guess and requires significant optimization but is nonetheless suitable for adaptation to upcoming terrain. We demonstrate experiments using the ANYmal C quadruped, with fully onboard sensing and computation, to cross obstacles at moderate speeds using this technique. Our second approach uses latent-mode trajectory regression (LMTR) to imitate expert data - while avoiding invalid interpolations between distinct behaviors - such that minimal optimization is needed. This enables high-speed motions that make more expansive use of the robot's capabilities. We demonstrate it on flat ground with the real robot and provide numerical trials that progress toward deployment on terrain. These results illustrate a paradigm for advancing beyond short-horizon dynamic reactions, toward the type of intuitive and adaptive locomotion planning exhibited by animals and humans.