There are an increasing number of applications that require the end-effector of a robotic manipulator to be compliant in response to the intended force exerted by a human operator. In physical human-robot interaction, the stability and compliance are often affected in the presence of both human and an environment, especially during the transition from noncontact to contact. Unlike previous works, the physical human-robot interaction in this letter takes into consideration of the interaction of the human, robot, and a moving environment. To achieve this, this letter proposes a variable admittance control strategy with an adaptive controller to produce a time-varying stiffness interacting with the human and a moving environment simultaneously. The passivity of the system is guaranteed for time-varying admittance with human interaction coupled with a moving environment. Experiments are conducted to demonstrate the capability of the controller.
11 March 2020 in Journal of Intelligent & Robotic Systems
There has been an increasing demand for physical human-robot collaboration during the design prototyping phase. For example, users would like to maneuver the end-effector compliantly in free space followed by supplying a contact force to obtain a firm adhesive connection. The technical challenges is the design of the controller, especially during the switching from human-robot interaction (human guides robot) to robot-environment interaction for the robot to continuously maintain the contact force even after the human lets go. Traditional controllers often result in unstable interaction during the switches of the controllers. Therefore, this paper proposes a control scheme that unifies impedance and admittance in the outer loop, and unifies the adaptive position and velocity control in the inner loop to address this issue. The cooperation of the cobot is divided into two modes, namely, an augmentation mode where the human force is the desired input to guide the motion of the cobot, and an autonomous mode where predefined position and force commands are used (e.g., to maintain a desired holding force). With the proposed control scheme, the physical interaction between the robot, human and environment can be smoothly and stably transited from augmentation mode to autonomous mode. Experiments are then conducted to validate the proposed approach.
28 March 2019 in Science China Information Sciences
There has been an increasing demand to automate the non-patient care matters so that the clinical staff can focus on delivering patient care. For example, out-patients undergoing chemotherapy increases their toilet usage frequency due to the treatment. As they are undergoing chemotherapy, their output waste contains a level of chemical. This task is compulsory yet troublesome and time-consuming so it is often desired to be removed from the nursing staff for them to focus on patient care. Hence, in this paper, we propose a manipulator system to automatically dispose the bedpan used by patients undergoing chemotherapy. The main technical challenge lies in the removal of the bedpan from the commode as the interaction of the grasping is highly dynamic, along with the different conditions of the bedpans. To address this manipulation issue, a Residual Reinforcement Learning (RRL) method that leverages vision-based commode pose estimation and the reinforcement learning (RL)-based uncertainty compensation for improvement of the grasping accuracy is proposed to increase the robustness of the disposal. The experiments conducted show that the manipulator can dispose the bedpan without human intervention and the proposed method achieves a 100 % success rate while the traditional method without RL is only 50 %.
13 June 2022 in IEEE Robotics & Automation Magazine