An Obstacle Avoidance Two-Wheeled Self-Balancing Robot

Abstract

An inverted pendulum is an inherently unstable system and cannot be maintained in a balancing state without control. There are several types of inverted pendulums, such as truck-type inverted pendulums, wheel-type inverted pendulums and rotary-type inverted pendulums. In this paper, we focus on a two-wheeled self-balancing robot (TWSBR) based on the wheel-type inverted pendulum. TWSBR can be controlled by many different methods. The most common method to control TWSBR is to use PID controller [1]. In [2], state feedback control by Linear Quadratic Regulator (LQR) is proposed. It is common that microprocessors such as ARM Cortex-M4 [3], STM32 [4] and AVR [5], [6] have been used to control TWSBR. Also, obstacle avoidance of robots has been researched for many years and many methods are adopted fuzzy inference. In [7], obstacle avoidance by multiple ultrasonic sensors for TWSBR is proposed.