Variable Hydraulic Transmissions for Passive Wearable Robots

Abstract

Wearable robots are used to assist or perform human-like motions. A teleoperated robot is used to transmit the motion from a specific point to another. Teleportation can be delivered through a passive or active system. Body-powered wearable robots (BPWRs) use passive teleoperated manipulation to transmit the power between different limbs where the power transmission between limbs needs to be at the top speed and force efficiency. Rehabilitation is one of the applications of BPWRs since they promise a portable, low-power, and safe interface. The active rehabilitation robots suffer from poor patient engagement due to motor slacking, where the user relies mostly on the motor to accomplish the task. BPWR is a passive alternative to let the users feel more active by using their own bodies to power the system. BPWR can also be used for surgical, performance amplifying, and puppeteering purposes. Hydraulics is a viable choice for wearable robots due to its high stiffness, light weight, and capability of having a variable transmission ratio. Hydraulic components usually operate inefficiently at low pressures due to viscous losses. Due to the safety concerns, low-pressure hydraulics is preferable in applications where the device has interaction with the human body. Efficient components and analytical models to describe their performance are needed to be incorporated in these systems. In this work, I present highly efficient and consistent components to be used in a wearable device. Furthermore, I propose analytical models to facilitate a better understanding of the components. A variable hydraulic transmission design is presented to provide a high number of unique transmission ratios. The valves in the system set a configuration in actuators' connectivity and vary the transmission ratio between joints. The proposed design requires low energy to switch the transmission ratios in a compact and wearable embodiment. I introduce a model for losses across the transmission to allow system-level understanding of the system. A superposition of mechanical and hydraulic losses is used to find the loss in the system with a hydraulic analogy. Models described in this work are validated with experimental tests. Soft robots are embodied to an antagonist transmission due to their conformity. A novel bone-inspired bending soft robot is introduced to be used in a teleoperated antagonist transmission setup. Higher stiffness and natural frequency are achieved in the bone-inspired bending soft robots compared to traditional bending soft robots. I propose experimentally validated dynamic and geometrical models to understand the stiffness and natural frequency of the bending soft robots and to predict the robot's motion trajectory.