Environment-adaptive SLAM (Simultaneous Localization and Mapping) rear-end FPGA (Field Programmable Gate Array) accelerator and design method

Abstract

The invention discloses an environment-adaptive SLAM (Simultaneous Localization and Mapping) rear-end FPGA (Field Programmable Gate Array) accelerator structure and a design method, and the accelerator structure comprises a configurable circuit module, an environment-adaptive parameter updating module and a gated clock which are arranged at the rear end of an SLAM, wherein the visual information data is divided into two paths through the SLAM front end; one path inputs visual information extracted from the environment into a configurable circuit module arranged at the SLAM rear end; the other path sends the number of the road sign points to an environment adaptive parameter updating module arranged at the rear end of the SLAM; controlling the number of the configurable circuit modules opened by using a gated clock circuit; moreover, according to the number of the road sign points collected by the SLAM front end, the configuration parameters enabling the power consumption of the configurable circuit module to be minimum in the current environment are solved, and the configuration parameters of the configurable circuit module are updated. Compared with the prior art, the architecture dynamic adjustment is realized, and meanwhile, the power consumption during operation can be reduced.

Description

technical field [0001] The invention belongs to the technical field of FPGA devices, and particularly relates to a design of an FPGA accelerator for the back end of a SLAM. Background technique [0002] SLAM (Simultaneous Localization And Mapping) is a key technology for drones and unmanned vehicles to determine their own motion trajectories without predicting map information. The SLAM system is divided into front-end and back-end. The front-end calculates the initial trajectory according to the collected sensor data. Due to the noise in the data collected by the sensor, the accuracy of the initial trajectory cannot meet the actual requirements. The back-end optimizes the initial trajectory calculated by the front-end. [0003] The SLAM backend usually uses a nonlinear optimization method to optimize the state. This method uses the error of the sensor observation data to construct a cost function, and obtains the motion state of the system by minimizing the cost function. T...

AI Technical Summary

Problems solved by technology

In practical applications, as the environment continues to change, the amount of data collected by the front-end of SLAM also changes, resulting in changes in the amount of calculations at the back-end of SLAM

In order to ensure the real-time performance of the calculation, the traditional SLAM back-end hardware accelerator is designed for the situation with the largest amount of calculation, and the hardware architecture is fixed during operation, so it cannot reduce power consumption according to changes in the environment

Images