Cruise Control Method for Intelligent Vehicles Based on Reinforcement Learning

This paper explores the application of Q-learning in Adaptive Cruise Control (ACC). Environment Modeling and Decision-Making: The radar detects signals from obstacles ahead and transmits them to the ACC system. The Q-learning algorithm makes decisions based on environmental changes, such as the speed and distance of the vehicle in front. Through trial and error and continuous learning, it optimizes the following strategy. Reinforcement learning is a computational method that automates decision-making and learning processes toward goal-oriented tasks. It is framed within the Markov Decision Process (MDP), using states, actions, and rewards to define the interaction between a learning agent and the environment. Reinforcement learning emphasizes the agents interactive learning with the environment, without the need for imitative supervision signals or complete environment modeling. The Markov Decision Process provides a theoretical framework for achieving goals through interactive learning. All reinforcement learning problems are considered an MDP process, serving as the fundamental framework for modeling reinforcement learning. By updating the state-action value function Q(s, a), the Q-learning algorithm selects actions in each state that maximize long-term returns, enabling intelligent control. Experimental results show that the method applied in this paper ensures driving safety and comfort, and it can be practically implemented.