A HW/SW Implementation on FPGA of Absolute Robot Localization Using Webcam Data

Abstract

This paper presents an implementation of absolute robot localization algorithm using FPGA technology. The adopted localization method uses webcam tracking images. This technique has been developed and implemented for the motion of the robot from an initial position towards another desired position. Firstly, we have validated the proposed approach on PC platform using C language and Open CV library. Secondly, to facilitate robot autonomous navigation, a mixed HW/SW implementation was been developed using a high performance version of the NIOS processor coupled with a custom hardware accelerator. Experimental tests on Altera Cyclone III FPGA Starter Kit show an improvement of 85% over purely software execution which proves the effectiveness of this proposed architecture.