Multi-sensor based object grasping with eye-in hand laser ranger

Abstract

During the last decade, industrial robots have been applied to various field of science from computer-assisted surgery to space exploration. In recent years, autonomous service robots are increasingly finding their way into our daily lives. In this paper, we present the design of the robot, and the robot arm path planning combined with several sensors: motor encoder, CCD camera and laser range finder. Using information of CCD-cameras that provide information of the robot's environment which serves as an input for a robust object recognition algorithm: SURF (Speeded Up Robust Features). However, image information provided by cameras is strongly affected by environmental influence. In order to reduce environmental influence, we integrate distance information provided from a laser range finder. The integrated sensory system locates the object using distance information and then identifies the target using image information. Ultimately the dexterous manipulator system integrated with visual servo system and eye-in-hand laser scanner feedback, it can perform object fetching towards contributions in flexible automation.