Biologically inspired control of a kinematic chain using the superposition of motion primitives

This paper concerns the control of a nonlinear system using a control paradigm inspired by recent experiments on the central nervous system of frogs and rats. This paradigm consists in expressing the control variable as the linear combination of a finite number of basic controllers, called motion primitives; the system is then controlled through the combinators. In this paper we use this biologically inspired paradigm for controlling a kinematic chain. The proposed motion primitives have two appealing features. First, they can be easily scaled to drive the system to a desired final state in an arbitrary time. Second, they are invariant under modifications of some geometrical parameters of the kinematic chain to be controlled. Finally, all these results are used for simulating human reaching movements; simulated trajectories are compared with human captured data.