Adaptive Control Method for Driverless Cars Considering Probabilistic Graph Path Planning for Intelligent Vehicles

Traditional path planning methods usually aim to optimize the shortest path length or time, which makes it difficult to accurately plan safe and feasible vehicle driving paths in complex environments. To solve this problem, the algorithm combines a potential field model and a probability graph algorithm. Under the premise of comprehensive evaluation of vehicle driving risks, an artificial potential field algorithm is used to construct a potential field model for the environment. This model mainly considers the coupling effects of multiple obstacles, environmental threats, and road conditions in the environment, and can accurately represent the interaction between vehicles and the environment. At the same time, probability graph algorithm is used for path planning to optimize the multi-dimensional transportation costs between nodes. This method establishes a multidimensional access cost evaluation matrix between sampling points, integrating multiple aspects of vehicle driving, such as road conditions, traffic flow, and turning radius. Finally, the simulation results indicate that the algorithm comprehensively considers the coupling effects of obstacles, environmental threats, and road conditions in the path planning process. The probability graph algorithm has been used to optimize multidimensional travel costs, which can generate feasible, safe, and efficient travel paths, providing a multi-objective optimization path planning algorithm for intelligent vehicles. The application of this algorithm can reduce the load of intelligent vehicle path planning, effectively solve path planning problems in complex environments, and further promote the development of unmanned intelligent vehicle technology.