Mobile Robot Path-Tracking Using an Adaptive Critic Learning PD Controller
This paper proposes a novel self-learning PD (Proportional- Derivative) control method for mobile robot path-tracking problems. In the self-learning PD control method, a reinforcement-learning (RL) module is used to automatically fine-tune the PD coefficients with only evaluative feedback. The optimization of the PD coefficients is modeled as a Markov decision problem (MDP) with continuous state space. Using an improved AHC (Adaptive Heuristic Critic) learning control method based on recursive least-squares algorithms, the near-optimal control policies of the MDP are approximated efficiently. Besides its simplicity, the self-learning PD controller can be adaptive to uncertainties in the environment as well as the mobile robot dynamics. Simulation and experimental results on a real mobile robot illustrate the effectiveness of the proposed method.
KeywordsMobile Robot Reward Function Critic Network Continuous State Space Adaptive Critic
Unable to display preview. Download preview PDF.
- 1.Choi, S.B.: The Design of a Look-Down Feedback Adaptive Controller for the Lateral Control of Front-Wheel-Steering Autonomous Highway Vehicles. In: Proceedings of the American Control Conference, Albuquerque, New Mexico, vol. 3, pp. 1603–1607 (1997)Google Scholar
- 3.Aguilar, M., et al.: Robust Path Following Control with Exponential Stability for Mobile Robots. In: IEEE International Conference on Robotics and Automation, Leuven, Belgium, vol. 4, pp. 3279–3284 (1998)Google Scholar
- 5.Sanchez, O., Ollero, A., Heredia, G.: Adaptive Fuzzy Control for Automatic Path Tracking of Outdoor Mobile Robots: Application to Romeo 3R. In: IEEE International Conference on Fuzzy Systems, Barcelona, Spain, vol. 1, pp. 593–599 (1997)Google Scholar
- 6.Kaelbling, L.P., et al.: Reinforcement Learning: a Survey. Journal of Artificial Intelligence Research 12, 237–285 (1996)Google Scholar
- 8.Lee, C.H., Teng, C.C.: Tuning of PID Controllers for Stable and Unstable Processes with Specifications on Gain and Phase Margins. International Journal of Fuzzy Systems 3(1), 346–355 (2001)Google Scholar
- 11.Barto, A.G., Sutton, R.S., Anderson, C.W.: Neuron-like Adaptive Elements that can Solve Difficult Learning Control Problems. IEEE Transactions on System, Man and Cybernetics 13, 834–846 (1983)Google Scholar
- 13.Sutton, R.: Learning to Predict by the Method of Temporal Differences. Machine Learning 3(1), 9–44 (1988)Google Scholar