Integrated planning and control of robot motion tasks.
Enhanced reactive capabilities.
Perform complex robot motion tasks.
 
 


Indirect-Mapping EKM
      
Use Extended Kohonen Map (EKM) to map input sensory space indirectly to ouptut motor control
       space.

                        

Cooperative EKMs
      
EKMs for target localization, obstacle localization, robot kin localization, motor control.

Online Self-organization of EKMs
      
EKMs self-organize to map sensory inputs to motor outputs.
       Result of self-organization:
                                                

    
 

 

Negotiating static obstacle
Negotiating moving obstacle
Motion in a changing environment
Motion in a complex environment
Motion in a complex environment
Cooperative tracking of targets
 

  A/Prof. Leow Wee Kheng, Dept. of Computer Science, National University of Singapore.
A/Prof. Marcelo Ang, Dept. of Mechanical Engineering, National University of Singapore.
Mr. Bryan Low Kian Hsiang (M.Sc.), Dept. of Computer Science, National University of Singapore.
 
  K. H. Low, W. K. Leow, and M. H. Ang, Jr. Action selection for single- and multi-robot tasks using
       cooperative extended Kohonen maps. In Proc. Int. Joint Conf on Artificial Intelligence, 2003,
       p. 1505-1506.

K. H. Low, W. K. Leow, and M. H. Ang, Jr. Enhancing the reactive capabilities of integrated planning
       and control with cooperative extended Kohonen maps. In Proc. IEEE Int. Conf. on Robotics and
       Automation
, 2003, p. 3428-3433.

K. H. Low, W. K. Leow, and M. H. Ang, Jr. Action selection in continuous state and action spaces by
       cooperation and competition of extended Kohonen maps. In Proc. Int. Joint Conf. on Autonomous
       Agent & Multi Agent Systems
, 2003, p. 1056-1057.

K. H. Low, W. K. Leow, and M. H. Ang, Jr. A hybrid mobile robot architecture with integrated planning
       and control. In Proc. Int. Joint Conf. on Autonomous Agent & Multi Agent Systems, 2002, p. 219-226.

K. H. Low, W. K. Leow, and M. H. Ang, Jr. Integrated planning and control of mobile robot with self-
       organizing neural network. In Proc. IEEE Int. Conf. on Robotics and Automation, 2002, p. 3870-3875.

 
  This project supported by NUS ARF R-252-000-018-112.

Updated 21 Nov 2003