Force compliant trajectory optimization

Force compliant trajectory optimization is a term used here to describe a method of following an optimized path while interacting with an environment having only partial a priori knowledge. This can be used when the start and goal poses are fixed and what knowledge is available allows an initial path to be given. A robot traversing this path then senses the interaction force with the environment and builds up a new optimized path during motion. The methods of potential fields are combined with the measured force gradient to allow the trajectory to change and place constraints on the workspace. The concept of a null field is introduced to negate any trajectory modifications at goal locations to accurately achieve positioning. The method is defined for an n-degree system and results are given for two and three degree of freedom serial robot simulations.