This paper presents a position control scheme for flexible joint robot based on online gravity compensation (OGC). After concluding some drawbacks of OGC-based PD control law, a nonlinear state feedback controller is designed, where analysis for overshoots of motor side and residual vibration of link side is provided. It is shown that asymptotic stability is guaranteed, and each signal of closed loop system is proved to be bounded via Lyapunov method. Comparing experiment results illustrate effectiveness of the proposed approach.

A novel reconfigurable manipulator, which is composed of active rigid joints and deformable links, can obtain relatively dexterous end-effector and extended workspace with fewer joints by bending its deformable links. However, frequent changes in links' shape bring difficulties to the model and control of the manipulator. Considering the typical hybrid behaviors caused by bending operation, a hybrid model and a model-free control framework with no prior kinematic information are proposed to describe and control the reconfigurable manipulator. The model-free control framework is based on the hypothesis that the Jacobian matrix is constant in a local region. The Jacobian matrix can be estimated by moving each actuator independently by an incremental amount and observing their effects on the robot's end-effector. The model and controller have been implemented on a 4-DOF reconfigurable manipulator with two deformable links. The experimental results validate our proposed hybrid model and model-free control framework. Our method can automatically adapt to kinematic changes, which is very difficult for model-based methods. Moreover, our method can fulfil the tracking task even if the given target is not in the current workspace.

An asymmetric three-degree-of-freedom parallel mechanism is adopted in rehabilitation robots for assisting patients suffering from stroke or trauma in the hip. It is necessary to keep its kinematic singularity out of the workspace of human normal gait and increase the output power efficiency. Therefore, a novel method is proposed to optimize geometrical parameters of the mechanism. To describe the kinematic singularity in a better way, the improved force transmission indexes based on previous methods are proposed using the reciprocal product and mobility condition of the closed-loop mechanism. The indexes mainly represent the force transmission performance of unactuated parts of subchains and moving platform. Together with the driving force transmission indexes and geometrical constraints, the multiobjective optimization model is established. The differential evolution algorithm, which is widely applied to mechanism optimization, is used to achieve optimal results. The Jacobian matrix singularity and output power efficiency along giving trajectory before and after optimization are compared to verify the effectiveness of the method.

Orientation control is an important topic with many applications including robotic manipulation, spacecraft control, etc. The orientation control is particularly challenging when the system model is unknown due to the fact that the rotation group is a nonlinear manifold. In this letter, we propose a kinematic-free orientation control method for a deformable manipulator whose kinematic parameters are unknown. We estimate the Jacobian matrix by independently and incrementally actuating each joint and observing their effects on the orientation of the affected end-effector. Considering the nonlinear property, we generate a temporary and anticipant target by cutting a small distance on the geodesic in each control period, instead of using the target orientation directly. The geodesic is from the current orientation to the target orientation in SO (3). Our method guarantees global convergence and can automatically adapt to drastic changes in kinematic parameters, which is very difficult for model-based methods. Our method is implemented to control a 4-DOF deformable manipulator. The simulation and experimental results demonstrate the effectiveness of our method.

To realize ideal force/torque source in applications such as monopod hopping robots and assistive robots, force/torque controlled series elastic actuators(SEA) are widely used components. Most of existing SEA force/torque control approaches are based on the simple SEA whose force/torque output is proportional to the position of the motor with respect to the payload, however the structure of SEAs is becoming more and more complex, which leads to a nonlinear trend in SEA models. To overcome the problem, we propose a general control method, which is not restricted by the SEA mechanical structure, to control SEA output torque based on nonlinear disturbance observer(NDOB). Specifically, the presented strategy yields the first control solution, with guaranteed theoretical analysis, to successfully address the nonlinear SEA torque control problem, with simultaneous payload angular velocity identification as an additional benefit. The tracking performance of the designed control system is theoretically ensured by Lyapunov analysis. Furthermore, taking a nonlinear SEA for instance, simulation and experimental results suggest the effectiveness and superior performance of the proposed method by comparing it with existing method.