Design and implementation of autonomous Lawn-Mower Robot controller

Abstract

The objective of this paper is to design and implement a Behavior-Based Lawn Mower Robot controller that can be used to mow grass from lawns and play grounds autonomously. The controller uses “sense-act” approach to work in dynamic, unstructured and unknown environment without having any reliance on surrounding world information. The controller is implemented using Motor Schema architecture, which uses continuous response encoding and cooperative coordination method for behavior coordination. A set of concurrently running behaviors are defined to perform mowing operation. Sonar ranging is used to detect and avoid obstacles. Shaft and visual odometry in coordinated form are exploited for local positioning, while Global Positioning System (GPS) is used for global positioning. Camera is used to detect grass field and optocouple sensors are utilized to differentiate between mown and un-mown grass.