Rastegarpanah, Alireza; (2017) Autonomous vision-guided bi-manual grasping and manipulation. In: Proceedings of Advanced Robotics and its Social Impacts (ARSO). IEEE Xplore: Austin, TX, USA.

Preview

Text Rastegarpanah.pdf - Accepted Version Download (1MB) | Preview

Abstract

This paper describes the implementation, demonstration and evaluation of a variety of autonomous, vision-guided manipulation capabilities, using a dual-arm Baxter robot. Initially, symmetric coordinated bi-manual manipulation based on kinematic tracking algorithm was implemented on the robot to enable a master-slave manipulation system. We demonstrate the efficacy of this approach with a human-robot collaboration experiment, where a human operator moves the master arm along arbitrary trajectories and the slave arm automatically follows the master arm while maintaining a constant relative pose between the two end-effectors. Next, this concept was extended to perform dual-arm manipulation without human intervention. To this extent, an image-based visual servoing scheme has been developed to control the motion of arms for positioning them at a desired grasp locations. Next we combine this with a dynamic position controller to move the grasped object using both arms in a prescribed trajectory. The presented approach has been validated by performing numerous symmetric and asymmetric bi-manual manipulations at different conditions. Our experiments demonstrated 80% success rate in performing the symmetric dual-arm manipulation tasks; and 73% success rate in performing asymmetric dualarm manipulation tasks.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item