Unified Contact Model and Hybrid Motion/force Control for Teleoperated Manipulation in Unknown Environments

Teleoperated manipulation under the human intelligence is an effective solution to confront complicated tasks in unknown environments. However, the uncertain contact in environments is the main challenge to achieve good teleoperation, and some inevitable issues such as nonlinearities, various uncertainties, constraints, and communication delays in local and remote robots should also be taken into account. In this article, a unified contact model is proposed to be the targeted environment interaction for remote robot, which can cover various conditions such as free motion and rigid contact by setting different model parameters. Subsequently, a hybrid motion/force controller is developed to cope with nonlinearities and various uncertainties by adaptive robust technique, thus guaranteeing system stability and good transient convergence to the unified contact model. Particularly, to handle misteleoperation or sudden change of conditions described in unified contact model, a model predictive control-based contact optimization method is developed to be the outer loop of hybrid motion/force controller, which plans the desired motion and force trajectories that meet the state and targeted interaction constraints. By the estimation and transmission of environment parameter, the environment dynamics is reconstructed in the local side, which provides the effective force feedback of remote contact environment for human operator. Since the signal transmitted is replaced by the estimated environment parameter, the power-cycle in communication channel is eliminated. This design can avoid the passivity stability problem caused by communication delays, so the stability and good transparency under delays can be guaranteed by the separate design of local and remote controllers. The comparative experiments are implemented and verify the effectiveness of proposed framework for teleoperated manipulation.