One promising approach for robots efficiently learning skills is to learn manipulation skills from human tutors by demonstration and then generalize these learned skills to complete new tasks. Traditional learning and generalisation methods, however, have not well considered human impedance features, which makes the skills less human-like and restricted in physical human-robot interaction (pHRI) scenarios. In particular, stiffness generalization has not been well considered. This paper develops a framework that enables the robot to learn both movement and stiffness features from the human tutor. To this end, the upper limb muscle activities of the human tutor are monitored to extract variable stiffness in real time, and the estimated human arm endpoint stiffness is properly mapped into the robot impedance controller. Then, a dynamical movement primitives (DMPs) model is extended and employed to simultaneously encode the movement trajectories and the stiffness profiles. In this way, both position trajectory and stiffness profile are considered for robot motion control in this paper to realize a more complete skill transfer process. More importantly, stiffness generalization and movement generalization can be efficiently realised by the proposed framework. Experimental tests have been performed on a dual-arm Baxter robot to verify the effectiveness of the proposed method.

In this paper, a reduced-complexity model of the human arm endpoint stiffness is introduced and experimentally evaluated for the teleimpedance control of a compliant robotic arm. The modeling of the human arm endpoint stiffness behavior is inspired by human motor control principles on the predominant use of the arm configuration in directional adjustments of the endpoint stiffness profile, and the synergistic effect of muscular activations, which contributes to a coordinated modification of the task stiffness in all Cartesian directions. Calibration and identification of the model parameters are carried out experimentally, using perturbation-based arm endpoint stiffness measurements in different arm configurations and cocontraction levels of the chosen muscles. Consequently, the real-time model is used for the remote control of a compliant robotic arm while executing a drilling task, a representative example of tool use in environments with constraints and dynamic uncertainties. The results of this study illustrate that the proposed model enables the master to execute the remote task by modulation of the directions of the major axes of the endpoint stiffness ellipsoid and its volume using natural arm configurations and the cocontraction of the involved muscles, respectively.

Motion planning for an anthropomorphic arm is discussed in this paper. A three-level motion planning framework “joint space-movement primitive space-task space” is established by introducing movement primitives as the bridge connecting the task space and joint space. The proposed method cannot only control the motion process of an anthropomorphic arm, but also simplify the motion planning of complicated operation tasks. On the basis of this, a specific human arm triangle model is proposed as an instantiation of the abstract movement primitive to describe the motion state of the anthropomorphic arm. Through introducing the concept of working plane, the forward and inverse kinematics among joint space, human arm triangle space, and task space are derived by coordinate transformation and geometric analysis. And then, the joint trajectories of two fundamental movement primitives based on the human arm triangle, including motion of moving on the working plane and self-motion of switching working plane, are obtained by solving differential equations. Finally, the validation and feasibility of the proposed method are verified by one simulation comparison with a traditional method and two real experiments.

Typically, the fingertips of a dexterous robotic hand are designed to be rigid and equipped with a variety of sensors to provide tactile perception. However, this design scheme renders the grasping of small objects difficult, because rigid fingertips usually have a small contact area with the objects. In this article, we propose a novel fingertip design that combines soft tactile skin and a particle-jamming structure. Benefiting from the particle-jamming structure, the soft skin and fluid-state jamming particle material ensure that the fingertip encloses the object as much as possible when in contact with it. Then, the application of a vacuum turns the particle material into a high-stiffness solid-state, allowing it to hold objects. Soft tactile skin integrated with permanent magnetic particles estimates contact information between the fingertip and objects. Thus, utilizing the soft tactile skin and particle-jamming structure the fingertip can perceive and grasp objects simultaneously. The proposed soft fingertip module is fabricated by molding, which reduces the processing cycle time and cost. In addition, a series of experiments are conducted to demonstrate the performance of this novel fingertip design. The results validate that the proposed robotic fingertip can grasp tiny objects while perceiving contact position and force information.