Compliant control of variable stiffness actuator considering lower-limb robot applications

To realize different tasks in human-robotic interaction, various types of variable stiffness actuators (VSAs) have been and recently are investigated. Control of a “physical Human-Robot Interaction” can be achieved by the stiffness variation of the VSA. However, the mechanical design for stiffness variation may limit the capacity to achieve low output stiffness and fast stiffness variations. These limitations may become especially evident in assistance/rehabilitation robot applications. To overcome these limitations, the closed-loop impedance control scheme can be employed to achieve a programmable impedance range and an impedance variation speed. This control scheme has been widely applied on the fixed-compliance joint but lacks an implementation on the VSA joint because of its existing capacity to control the impedance with the mechanical construction. Hence, this dissertation contributes to providing a foundation for an impedance-controlled VSA within the context of lower-limb robot applications. The VSA application research of this dissertation is based on a self-developed prototype, i.e., the mechanical-rotary variable impedance actuator (MeRIA). The working principle of this prototype is based on controlling the effective length of bending bars (spring) for stiffness variation. The MeRIA can independently perform joint control and stiffness control through two independent-setup motors. A cascaded impedance control framework consisting of the position-torque-velocity control loops is designed for the joint control loop. Among these loops, the torque control loop is dominant since it mainly determines the performance of the torque-controlled actuator. Robustness plays a significant role in designing the torque-controlled compliant actuator. Therefore, a robust stabilization torque controller was achieved in the MeRIA system. However, the VSA plant is an inherently parameter-dependent system due to the controllable stiffness element. Since the VSA operates at a set of operating points, the aim is to achieve an adaptive control approach. Due to this, a gain-scheduled torque controller is proposed. The control performance is expected to recover robustness when stiffness values are varied from small to large. At the same time, the bandwidth is maximized, taking into account hardware limitations. In this way, a good trade-off between stability and performance can be achieved. A novel application of the VSA-based assistance/rehabilitation robot-featured impedance control is then proposed based on the above-mentioned torque control research. The robot follows the adaptive impedance control paradigm, thereby achieving an adaptive assistance level. In the proposed approach, the task performance during movement training can be improved with regard to 1) safety, e.g., when the subject intends to contribute considerable effort, low-gain impedance control is activated with a low-stiffness actuator to further decrease output impedance; and 2) tracking performance, e.g., for the subject with less effort, high-gain impedance control is used while pursuing high stiffness to enhance the torque bandwidth. Regarding the safety aspect, in this work, the torque controller designed at low stiffness is proven to be sensitive to disturbance for low output impedance while maintaining tracking performance. A precondition for this is to treat the input disturbance separately, which is guaranteed by the observer used in the torque control loop. Based on the proposed control approach, a robot with an impedance-controlled VSA joint can extend the capacity of bandwidth and low output impedance. This is an improvement compared to the impedance-controlled fixed-compliance joint. The effectiveness of all torque control schemes was experimentally verified using the MeRIA prototype. Moreover, the impedance-control experiments with a human test person were verified using the MeRIA-based one-degree-of-freedom lower-limb exoskeleton