Neuro-fuzzy algorithm implemented in Altera's FPGA for mobile robot's obstacle avoidance mission

Abstract

This paper presents the designed obstacle avoidance program for mobile robot that incorporates a neuro-fuzzy algorithm using Altera¿ Field Programmable Gate Array (FPGA) development DE2 board. The neuro-fuzzy-based-obstacle avoidance program is simulated and implemented on the hardware system using Altera Quartus <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> II design software, System-on-programmable-chip (SOPC) Builder, Nios <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> II Integrated Design Environment (IDE) software, and FPGA development and education board (DE2). Nios <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">®</sup> II IDE software is used to simulate and process the weight training. A mobile robot serves as the test platform of the program. An ultrasonic sensor and 3 servo motors are used as the test platform's sensing element and actuator, respectively with 2 servo motors used to move the robot and single servo motor used to adjust the sensing direction of the ultrasonic sensor. Altera¿ FPGA development board proves to be a successful embedded system platform for the neuro-fuzzy algorithm to be implemented and tested. Remote testing is done on mobile robot for obstacle avoidance missions via an established radio frequency (RF) communication using DIGI¿ XBee RF module.