79. Independent grasping scheme of space‑servicing‑oriented dexterous hand

Zhuang Peng1, Jinbao Chen2, Chen Wang3, Meng Chen4

1, 2, 3Nanjing University of Aeronautics and Astronautics, Nanjing, China

4Shanghai Institute of Aerospace Systems Engineering, Shanghai, China

1Corresponding author

E-mail: 1zhuanggeph@nuaa.edu.cn, 2chenjbao@nuaa.edu.cn, 3wangchen101@126.com, 4workmailcm@126.com

(Received 8 July 2015; received in revised form 24 August 2015; accepted 9 October 2015)

Abstract. It is difficult for the robot to grasp objects of any pose by the independent grasping scheme without the help of human. And the independent grasping scheme is also the key technology to develop the AI robot. In order to solve the problem, this paper establishes the full 3D point cloud model of the target object in advance under the PCL point cloud library. The partial view model of the target object in the current working environment is extracted when dexterous hand grasps the target. After aligning the extracted partial view model, the best alignment homogeneous transformation matrix mapping the partial view model to the full 3D point cloud model is obtained. According to the inverse matrix of the obtained matrix, the 3D point cloud model of the target object in the current working environment, where dexterous hand grasps the target, is obtained through the homogeneous transformation. According to the characteristics of the dexterous hand, a grasping algorithm is proposed, which is suitable for the most objects. Finally, the algorithm is verified by the point cloud model of the target object, and finds out the grasping points and grasping pose accurately (the direction of the hand's force). It demonstrates that this algorithm is correct and useful.

Keywords: 3D point cloud model, homogeneous transformation matrix, target alignment, independent grasping scheme, grasping point and pose.

References

[1]        Bluethmann W., Ambrose R., Diftler M., Askew S., et al. Robonaut: a robot designed to work with humans in space. Autonomous Robots, Vol. 14, Issues 2‑3, 2003, p. 179‑197.

[2]        Okita S. Y., Ng-Thow-Hing V., Sarvadevabhatla R. Learning together: ASIMO developing an interactive learning partnership with children. The 18th IEEE International Symposium on Robot and Human Interactive Communication, 2009, p. 1125‑1130.

[3]        Chen J. B., Han D., Peng Z. Active grasping control of virtual-dexterous-robot hand with open inventor. Mathematical Problems in Engineering, Vol. 65, Issue 1, 2014, p. 82‑87.

[4]        Wu Y. G., Qian J. W. A survey of grasp synthesis method of dexterous hand. China Mechanical Engineering, Vol. 15, Issue 3, 2004, p. 276‑281.

[5]        Nguyen V. D. Constructing force-closure grasps. International Journal of Robotics Research, Vol. 3, Issue 3, 1986, p. 1368‑1373.

[6]        Nguyen V. D. The synthesis of stable grasps in the plane. IEEE International Conference on Robotics and Automation, 1985, p. 884‑889.

[7]        Morales A., Asfour T., Azad P., et al. Integrated grasp planning and visual object localization for a humanoid robot with five-fingered hands. International Conference on Intelligent Robots and Systems, 2006, p. 5663‑5668.

[8]        Phoka T., Niparnan N., Sudsang A. Planning optimal force-closure grasps for curved objects by genetic algorithm. IEEE Conference on Robotics, Automation and Mechatronics, 2007, p. 1‑6.

[9]        Li J. W., Hu H. Y., Wang B., et al. Simple algorithm for internal force of multiple fingers grasping. Progress in Natural Science, Vol. 15, Issue 6, 2005, p. 733‑738.

[10]     Rusu R. B., Blodow N., Beetz M. Fast Point Feature Histograms (FPFH) for 3D registration. IEEE International Conference on Robotics and Automation, 2009, p. 3212‑3217.