Learning based control improves teamwork and safety in vehicle systems
Learning Agent-based Model Predictive Control for Holistic Vehicle Performance
Robotics
Summary
Controlling multiple parts of a vehicle to work together really well is tough, especially when some parts are hard to predict. The authors introduce a new system called learning agent-based model predictive control (LAMPC), which combines traditional control methods with learning from data to better predict unknown factors. This helps the vehicle perform better overall and keeps it safe, even when some behaviors haven't been learned well yet. They tested it through simulations and real experiments, showing it works better than previous control methods and can run quickly in real time.
agent-based model predictive controlmulti-agent systemsGaussian process regressionmodel predictive controlmachine learningstochastic controlchance constraintsreal-time controlvehicle performanceonline learning
Authors
Jiaming Zhong, Reza Valiollahi Mehrizi, Mohammad Pirani, Chao Yu, Alireza Kasaiezadeh, Yash Vardhan Pant, Amir Khajepour
Abstract
Agent-based model predictive control (AMPC) has recently been proposed as a distributed scheme that collaborates with all agents to achieve optimal holistic performance. However, its optimality highly depends on the prediction accuracy that requires all agents or their contributions to be known, which is too idealistic for actual implementation. This research proposes a novel practical hybrid control scheme - learning agent-based MPC (LAMPC), combining the model-based AMPC approach and data-based learning methods to improve the holistic vehicle performance for multi-agent systems. The Gaussian process regression (GPR) enhanced by an online data management strategy serves as the learning core to predict unknown contributions. A novel multi-step prediction mechanism leverages the GPR learning potential along the horizon. The predicted mean, representing the learned unknown contributions, completes the system model in the MPC for more accurate control. Meanwhile, a stochastic framework is formulated to guarantee control safety and feasibility using soft chance constraints based on the prediction variance. Both simulations and experiments show that, with the learning capability, LAMPC outperforms the traditional AMPC. LAMPC can achieve higher tracking performance in well-learned scenarios and always guarantee constraint satisfaction even in less-learned scenarios. Moreover, the proposed hybrid control scheme is efficient for real-time implementation and is flexible to any control agent topology.