Research On Key Technologies Of Autonomous Unmanned Vehicle System Based On Path Planning And LDPC Code

In recent years,autonomous driving technology has become a research hotspot in the field of transportation.Currently,the unresolved problems in unmanned driving system technology include the reaction and avoidance of unmanned vehicles to obstacles in the path planning process,as well as the error control problem in the transmission of information.1.Safety issues: Unmanned driving often experiences collisions and getting lost due to complex road conditions.Path planning aims to improve the safety,reliability,and efficiency of vehicle navigation,and can prevent collisions between vehicles and obstacles or vehicles and vehicles during driving.2.Communication issues: In the process of path planning,unmanned vehicles need to obtain a large amount of information in real-time,including location,obstacle,traffic information,and vehicle data.And higher requirements proposed for the accuracy and efficiency of data transmission.Meanwhile,channel transmission is easily affected by external environmental factors,such as multipath effects and noise interference.This may cause misjudge during path planning and lead to serious traffic accidents.The main work of this paper is as follows:(1)Research on path planning.Considering the current research trends in path planning,this paper adopts the artificial potential field method as the algorithm scheme.This paper proposes a virtual potential field detection circle model in order to improve the “minimum trap” problem which caused by excessive repulsion or force balance.In addition,to optimize the virtual potential field detection circle model for dynamic environments,this paper proposes a structure combining the LSTM recurrent neural network with the Q-Learning reinforcement learning algorithm.The feasibility of the proposed algorithm is verified by using the MATLAB simulation tool.It proves that unmanned vehicles can effectively avoid obstacles and have good adaptability to dynamic environments.(2)Research on wireless communication.In order to improve the reliability and efficiency of information transmission during the path planning process,an error control code is introduced to wireless communication transmission,which can reduce the decoding error rate and enhance the system’s fault tolerance.This paper adopts the LDPC code,which has superior error correction performance,as the coding and decoding scheme for wireless communication.The message passing mechanism of the LLR BP decoding algorithm is improved,and layered decoding strategy decoding algorithm is proposed to accelerate the convergence speed of decoding.Experimental evidence shows that the decoding performance is improved by approximately 0.2d B.Then the binary LDPC code is extended to a multivariate LDPC code,which effectively improves the decoding performance.Experimental evidence shows that the decoding performance is improved by approximately0.2d B.Furthermore,to facilitate hardware implementation,a decoding algorithm solution with multiple simplified layered decoding strategy is proposed.Finally,the decoding performance before and after improvement is analyzed and compared through MATLAB simulation platform.(3)Aiming at the requirement of LDPC code as a wireless communication error control coding and decoding scheme,the improved LDPC decoding algorithm is implemented in hardware using FPGA and Verilog HDL.This involves designing a suitable decoder structure and conducting overall as well as modular design or debugging.Finally,the accuracy of the design is verified by the timing simulation diagram and decoding results using Vivado and Model Sim.