Implementation of sensory motor coordination for robotic grasping

In this paper, human motor learning model based sensory motor coordination (SMC) algorithm is implemented on robotic grasping task. Compare to conventional SMC models, which connect sensor to motor directly, the proposed method used biologically inspired human memory structure in conjunction with SMC algorithm for fast grasping force control of robot arm. To characterize various grasping objects, pressure sensors on hand gripper were used. Measured sensor data are fed to short-term memory (STM) to design motor plan promptly using direct connection architecture between sensor and motor, and single layered neural network was applied to mimic STM in human memory structure. Through motor learning procedure, successful information is transferred from STM to long-term memory (LTM). Experimental results showed that the proposed method can control the grasping force adaptable to various shapes and types of grasping objects, and also it showed quicker grasping-behavior learning time compare to simple feedback system.