Lyapunov-Based Adaptive Neural Network Optimized Backstepping Control of Uncertain Unmanned Fire Fighting Robot

This paper studies the Lyapunov-based adaptive neural network optimized tracking control problem for a class of unmanned fire fighting robots. Firstly, by reasonably simplifying the unmanned fire fighting robot (UFFR) and combining it with its actual working scene, a novel system model is created that takes into consideration both system uncertainties and external disturbances, including unknown friction factors and drag force. Then, the optimized tracking control scheme for the UFFR is devised by integrating both adaptive neural networks and the backstepping technique. The objective of introducing adaptive neural network technique is to overcome the challenge posed by solving the Hamilton-Jacobi-Bellman (HJB) equation. Based on Lyapunov stability theory, it is demonstrated that all signals in the closed-loop system are semi-globally ultimately bounded and the output variables follow the reference signals to the desired accuracy. In the end, to validate the effectiveness of our designed control scheme, numerical simulations and practical platform experiments have been conducted. To ensure repeatability, our codes are open sourced on Github: https://github.com/JiannanChen/RL-based-OBC-of-UFFR.git