Neuromorphic path generation in real-time

Abstract

The problem of real-time trajectory generation in the decentralized control of a manipulator is considered. The trajectory generator presented is based on a vector-integration-to-endpoint (VITE) model of neural dynamics in planned human arm movements. In contrast to approaches which require the stipulation of the desired individual joint positions, the trajectory generator operates with desired coordinates of the end effector and generates the individual joint driving functions in real time, using geometric constraints which characterize the manipulator. The resulting manipulator trajectory exhibited VITE model characteristics including synchronous joint operation. An illustrative two-joint manipulator example is given.<>