SLAM operation device and method

Abstract

An SLAM (simultaneous localization and mapping) hardware accelerator device comprises a storage part, an operation part and a control part, wherein the storage part is used for storing input data, temporary operation result data, final operation result data and an instruction set and / or algorithm parameter data needed in the operation process; the operation part is connected with the storage partand used for completing calculation on an SLAM relevant algorithm and an application; and the control part is connected with the storage part and the operation part and used for controlling and coordinating the storage part and the operation part. The invention furthermore provides a method for completing SLAM operation. According to the method, data transmission, data operation, program running and the like are controlled through instructions. Through the device and the method, the SLAM algorithm can be accelerated effectively according to different demands, operation meeting with different demands is realized, and the device and the method have the advantages of high flexibility, high configurable degree, high operation speed, low power consumption, etc.

Description

technical field [0001] The invention relates to a SLAM (simultaneous Localization and Mapping, real-time positioning and mapping) computing device and method, which are used to accelerate the computing of the SLAM algorithm according to different requirements. Background technique [0002] Autonomous navigation in unknown environments is a basic capability of mobile robots (such as unmanned ground and air vehicles, etc.). The real-time positioning and mapping positioning in the SLAM task mainly completes the determination of the position of the robot in the map, and the main task of mapping is for the robot to establish a map of the corresponding environment according to the environment. In the absence of an initial map of the location environment, this requires the robot to be able to build a map in real time and use the map to complete its own positioning, and the SLAM algorithm required to complete this task is generated accordingly. However, accurately implementing SLAM...

AI Technical Summary

Problems solved by technology

Although this method is a device specially used to perform graphics and image operations, due to the complexity of SLAM algorithm operations, this method cannot well support its subsequent operations, that is, it cannot effectively accelerate the overall operation of SLAM algorithms.

At the same time, the GPU on-chip cache is too small to meet the computing needs of a large number of SLAM algorithms

In addition, since it is difficult to transplant similar structures such as CPU or GPU to robots in the field of practical applications, there is no dedicated SLAM hardware accelerator structure with strong practicability and high flexibility.

Images