Upper Limb Flexion Assistance Based on Minimum-Jerk Trajectory Using Wearable Motion-Assist Robot

In this paper, we propose a robotic system to assist patients who have upper limb dysfunction in performing reaching movements through flexion. Since upper-limb motion is more strongly needed than lower limb mobility for near work, a patient’s level of recovery of upper-limb function influences daily life. Recently, with the widespread application of robotic technology in rehabilitation medicine, it is often noted that moving actively is more important than moving passively to enable rapid recovery. Human reaching movement is known to conform to the standard minimum-jerk, which is characterized by a bell-shaped velocity waveform with a single peak. Patients with dysfunction move their extremities if they are capable of controlling the motor movement of their upper limb, but are assisted by the robot when they cannot do so. The range of movement is estimated from the motor control based on minimum-jerk criterion. We currently carry out research and development of various assistive robots for upper limb movements including power assist robots to supplement the shortage of medical practitioners and caregivers (Tsukahara et al. [2009]). If a robot can instead assist the patient, home rehabilitation is possible. Robots support those with dysfunction of the upper limb which leads to rehabilitation, improved quality of life, and reduced burden on medical practitioners and caregivers. Many rehabilitation robots have been developed, but they are implemented with a joystick to compensate for reduced limb function. In addition, there are a few master-slave robots, but these have shortcomings that limit practicality and independence of patients. Also, many power assist robots assist patients on the basis of pressure sensor data obtained from the physical operating forces. Most robots, however, utilize systems that are input amplifier based and cannot be used by people who do not have the range of motion (ROM) to exert sufficient force. Therefore, we developed an assistive robot for upper limb movement that has high rehabilitation effectiveness. By using a robot in daily life, patients can recover from dysfunction. Our robot enables flexion and extension of ⋆ This work was supported by Regional Innovation Cluster Program (City Area Type): Southern Gifu, Japan Area. the elbow by providing reaching movement support that takes into account the expanding range of motion of a joint. This reaching movement support method is based on estimating the trajectory of the participant’s reaching movement. 1. ASSISTIVE ROBOT FOR UPPER LIMB MOVEMENT The assistive robot for upper limb movement developed in this study is shown in Fig. 1, and its link structure is shown in Fig. 2. This robot has a manipulator with 4 degrees of freedom, and supports flexion and extension of the elbow. The drive axis is only in the first joint of the robot, while the second joint is a free joint for medial and external elbow rotation. The third joint is a prismatic joint with the flexibility to meet individual forearm-length. The forth joint is a free joint and is used for pronation and supination of the wrist. These free joints contribute to the free motion of the upper limb. Our motion assist robot is equipped with a DC motor that has a maximum torque and speed of 4.2[Nm] and 226[deg/s], respectively. These values are based on the results of a preliminary experiment examining elbow joint torque and operation speed (elbow joint torque: 4[Nm], operation speed: 150[deg/s]). The Preprints of the 18th IFAC World Congress Milano (Italy) August 28 September 2, 2011 Copyright by the International Federation of Automatic Control (IFAC) 5962 exterior of the robot and upper limb guard are made of acrylonitrile butadienestyrene resin (ABS resin) to reduce the robot’s weight. A super-thin film pressure sensor detects the force applied by the user. Four sensors are installed on the robot and measure the forces of flexion and extension, as well as medial and external rotation of the elbow by determining the fixed pressure between the robot and the human arm. The robot is attached to the upper limb by a shoulder supporter (see Table 1 for the robot’s specifications). Fig. 1. Upper limb motion assist robot