Rocket Landing Control with Random Annealing Jump Start Reinforcement Learning

TL;DR

This paper introduces a novel reinforcement learning approach called Random Annealing Jump Start (RAJS) that significantly improves rocket landing success rates from 8% to 97% by leveraging prior feedback controllers and advanced exploration strategies.

Contribution

The paper presents RAJS, a new RL method that enhances exploration and learning efficiency for goal-oriented rocket landing control, with proven high success rates and real-time applicability.

Findings

Success rate increased from 8% to 97%.

Validated through extensive simulation and Hardware-in-the-Loop testing.

Enhanced exploration reduces training time and improves policy robustness.

Abstract

Rocket recycling is a crucial pursuit in aerospace technology, aimed at reducing costs and environmental impact in space exploration. The primary focus centers on rocket landing control, involving the guidance of a nonlinear underactuated rocket with limited fuel in real-time. This challenging task prompts the application of reinforcement learning (RL), yet goal-oriented nature of the problem poses difficulties for standard RL algorithms due to the absence of intermediate reward signals. This paper, for the first time, significantly elevates the success rate of rocket landing control from 8% with a baseline controller to 97% on a high-fidelity rocket model using RL. Our approach, called Random Annealing Jump Start (RAJS), is tailored for real-world goal-oriented problems by leveraging prior feedback controllers as guide policy to facilitate environmental exploration and policy learning in RL. In each episode, the guide policy navigates the environment for the guide horizon, followed by the exploration policy taking charge to complete remaining steps. This jump-start strategy prunes exploration space, rendering the problem more tractable to RL algorithms. The guide horizon is sampled from a uniform distribution, with its upper bound annealing to zero based on performance metrics, mitigating distribution shift and mismatch issues in existing methods. Additional enhancements, including cascading jump start, refined reward and terminal condition, and action smoothness regulation, further improve policy performance and practical applicability. The proposed method is validated through extensive evaluation and Hardware-in-the-Loop testing, affirming the effectiveness, real-time feasibility, and smoothness of the proposed controller.