A vehicle system based on fpga design and a star path search method

Abstract

The invention discloses a vehicle-mounted system based on FPGA design and an A star path searching method. The vehicle-mounted system comprises a heterogeneous computing system composed of a vehicle-mounted computing platform for providing map information and an FPGA for computing a path. The FPGA design comprises the following modules: a map node information storage sub-module and a map node serial number storage sub-module used for storing map information, a node expansion module used for expanding nodes based on an evaluation function, a node information updating module used for calculatingthe relationship between the evaluation function of a newly expanded node and a father node, a first round of sorting module used for sorting newly expanded nodes based on the evaluation function value and an odd-even sorting network, and a second round of sorting module used for sorting all open nodes based on the evaluation function value and a doubly linked list. According to the method, the advantages of the FPGA in the aspect of parallel operation are brought into full play in FPGA design, the efficiency of path search can be greatly improved, and vehicle-mounted application is achievedbased on communication and system design.

Description

technical field [0001] The invention relates to the field of electronic technology, in particular to a vehicle-mounted system designed based on FPGA and an A star path search method. Background technique [0002] Intelligent networked vehicles are currently a hot spot in the development of the automotive industry, and have great potential in improving the safety, comfort, energy saving and efficiency of vehicles and traffic. In recent years, with the development of intelligent driving technology, increasingly complex intelligent driving algorithms have also put forward higher requirements for on-board computing hardware: under the premise of reasonable cost and energy consumption, the computing power of hardware should be improved to achieve higher Calculate the frequency and control accuracy, and apply more complex intelligent algorithms. [0003] Currently, mainstream processors include CPUs, GPUs, FPGAs, and ASICs. Among them, FPGA (Field Programmable Gate Array) has be...

AI Technical Summary

Problems solved by technology

The main process of the A star algorithm includes basic processes such as node expansion, evaluation, and sorting. The storage of map coordinate information and the sorting process of node evaluation function values ​​may increase nonlinearly with the improvement of accuracy, thereby increasing the calculation delay of the decision-making system. time, it is difficult to meet the real-time requirements of intelligent vehicle decision-making

Images