Modelling and Control of Hyper-Redundant Micromanipulators for Obstacle Avoidance in an Unstructured Environment

Abstract

The complexity of the obstacle avoidance problem in an unstructured environment increases when the task is performed in the micro scale, which necessitates the adoption of novel control techniques for obstacle inference and avoidance. In this article, we introduce a Van der Waals (VdW) force extension to the traditional dynamic micromanipulator model, where every link is decomposed into a series of elementary particles that interact with neighboring objects during the manipulator's motion. This interaction, along with the inherent nanoscale friction, introduce additive nonlinearities in the model that are compensated through an adaptive positioning control scheme. The estimation of the induced VdW forces are used to derive an approximation of the obstacle's position, followed by an obstacle avoidance algorithm, which generates a collision free path. Simulation studies on a hyper-redundant micromanipulator are offered to highlight the effectiveness of the proposed scheme.