Intelligent arm manipulation system in life science labs using H20 mobile robot and Kinect sensor

摘要

In service robot applications, the detection and pose calculation of the desired objects are very essential to manipulate them. This paper presents an intelligent system to realize manipulation tasks for different objects in life science laboratories. This system uses the H20 mobile robots equipped with Kinects as visual sensors to recognize the required target and identify its pose related to the robot body. The H20 robot has dual arms where each arm consists of 6 revolute joints with 6-DOF. For each arm, the closed-form solution for the inverse kinematic problem is found. Different grippers and handles which are attached to a labware container are designed for this purpose. The detection and localization methods are performed for single colored and textured targets related to the handle to be grasped and to the placing holder. The methods are based on HSV color segmentation and also on local features recognition with SURF (Speeded-Up Robust Features). Client-Server communication models are developed to connect the navigation control with the arm controller which is in turn also connected with the Kinect control for visual processing.