FPGA-based point cloud feature matching method and system and path planning system

Abstract

The invention provides a point cloud feature matching method and system based on an FPGA, and a path planning system. The system comprises a processing system part and a programmable logic part. The processing system part performs periodic point cloud data acquisition, the acquired point cloud data is written into a specified address space of a memory and communicates with the programmable logic part through a register channel, and the programmable logic part acquires the point cloud data from the specified address space frame by frame and performs logic processing. The KNN process of processing feature point cloud matching on a traditional processor CPU is avoided and simplified.

Description

technical field [0001] The invention relates to the field of data processing, in particular to an FPGA-based point cloud feature matching method, system and path planning system. Background technique [0002] With the rapid development of emerging technologies such as artificial intelligence and autonomous driving, lidar is widely used in the fields of artificial intelligence, intelligent robot perception, navigation, positioning, etc., and has achieved good results in low-to-medium speed mobile robot special operation scenarios . The requirements for the performance, size, power consumption, and cost of the application system are also getting higher and higher. At present, in the field of mobile robots, lidar has great advantages in environmental perception, positioning, and obstacle avoidance. However, the data processing flow of lidar is almost all processed on traditional industrial computers and CPU or GPU, including point Processes such as cloud data acquisition, fea...

AI Technical Summary

Problems solved by technology

[0005] The data processing process of industrial computer + CPU, because the CPU is a serial execution process, even a multi-core multi-threaded CPU, when processing massive data and business logic, is time-consuming and inefficient, and cannot meet the real-time requirements of the business

Moreover, the Moore's Law CPU, which is about to fail, can hardly continue to support the energy efficiency of complex data processing. The industrial computer is large in size, power consumption, and high cost cannot be ignored.

However, when the SIMD instruction architecture of the GPU is dealing with irregular and irregular data structures, the acceleration performance cannot be well exerted.

Images