The Research Of Apprenticeship Learning Algorithm Applied In The Unmanned Car High-speed Driving In The Simulated Environment

In recent years, with ongoing exploration and research of the field of artificial intelligence and machine learning, the application of machine learning methods in real life is also not uncommon, as the field of web search, machine vision, speech recognition and automatic driving.Traditional apprenticeship learning algorithm autopilot unmanned car in the high-speed environment analog behavior analysis, improved core part of the apprenticeship learning algorithm-the reward function feature weight vector selection algorithm, and in the initial policy choice on optimization, to improve the efficiency and accuracy of their in automotive automatic driving behavior simulation application. The main work of this paper the following three parts:(1) Propose an improved apprenticeship learning algorithm-maximum boundary method based apprenticeship learning algorithm aim to the traditional apprenticeship learning algorithm. Optimize the initial strategy selection process of the traditional apprenticeship learning algorithm;(2) Apply the apprenticeship learning algorithm to the unmanned car in the high-speed environment. Provide a new model and method for the unmanned car problem;(3) Verify the effectiveness of apprenticeship learning algorithm based on the maximum boundary method, and describe the future direction of the work..