Online Trajectory Planning Through Combined Trajectory Optimization and Function Approximation: Application to the Exoskeleton Atalante

TL;DR

This paper introduces Guided Trajectory Learning, an efficient online trajectory planning algorithm that learns from offline optimized solutions, enabling real-time control of robots like the Atalante exoskeleton without heavy computational demands.

Contribution

The paper presents a novel algorithm that combines trajectory optimization with function approximation for real-time robotic trajectory planning, applicable to various systems.

Findings

Achieves real-time trajectory generation for the Atalante exoskeleton.

Demonstrates computational efficiency and practicality for online robotic control.

Applicable to different robotic platforms with minimal setup.

Abstract

Autonomous robots require online trajectory planning capability to operate in the real world. Efficient offline trajectory planning methods already exist, but are computationally demanding, preventing their use online. In this paper, we present a novel algorithm called Guided Trajectory Learning that learns a function approximation of solutions computed through trajectory optimization while ensuring accurate and reliable predictions. This function approximation is then used online to generate trajectories. This algorithm is designed to be easy to implement, and practical since it does not require massive computing power. It is readily applicable to any robotics systems and effortless to set up on real hardware since robust control strategies are usually already available. We demonstrate the computational performance of our algorithm on flat-foot walking with the self-balanced exoskeleton Atalante.