Multi-path vehicle track information processing system based on millimeter wave radar group

Abstract

The invention relates to a multi-path vehicle track information processing system based on a millimeter wave radar group, which is characterized in that a vehicle signal detected by a radar is received by an antenna of the radar, reaches a radar CPU (Central Processing Unit) through an amplifier and a DSP (Digital Signal Processor), and generates a digital signal through a CAN (Controller Area Network) controller on the radar CPU; the digital signal is converted into a high-low level signal through a physical layer TJA1050 in the radar, the high-low level signal can be transmitted through a differential CAN line, and the high-low level signal is transmitted to an industrial personal computer; a link layer MCP2515 in the industrial personal computer transmits a radar signal to a Raspberry Pi bcm2711 through an SPI bus, and the signal is transmitted to a spi on a pcb through an SPI protocol; and a DMA directly accesses a memory to transmit information to an MEM memory, the CAN transmitting and receiving speed is increased, finally, signals of four paths of radars are accessed to four cores of ARM A72*4 through two paths of CAN lines and four paths of spi controllers, and four paths of radar information are synthesized. According to the system, the problem of high time delay of a distributed computing method is solved, the thread interruption frequency is greatly reduced, the CPU computing efficiency is improved, and larger data volume can be processed.

Description

technical field [0001] The invention relates to the field of intelligent traffic perception, in particular to a multi-channel vehicle track information processing system based on a millimeter wave radar group. Background technique [0002] In recent years, with the rapid economic and social development, as well as the advancement of roadside sensor technology and the decline in cost, vehicle trajectory perception technologies based on various roadside sensors have emerged in an endless stream. At present, the widely used vehicle trajectory recognition method at home and abroad is based on camera. The vehicle trajectory recognition method, on the one hand, the camera is not sensitive to the speed perception of the vehicle, therefore, the vehicle trajectory acquired by the cameras on different road sections cannot be effectively spliced ​​based on the vehicle motion characteristics; on the other hand, the recognition accuracy of the camera for the remote vehicle is low, Vehicl...

AI Technical Summary

Problems solved by technology

Solve the problem of high delay in distributed computing methods, and greatly reduce the frequency of thread interruption, improve CPU computing efficiency, and can handle larger amounts of data

Images