For years, crane, a chain block, an elevator and a forklift truck have been developed and used to carry heavy loads, but manpower needed where heavy equipment use is not practical. Aging workers suffer from musculoskeletal disorders, and are helped by developing various muscle assisting wearable robots. Industrial wearable robots must meet the payload capacity required for the pilot"s overall operation to ensure safety and operational performance. However, the payload capacity of wearable robot using rotary actuator or linear actuator at the knee joint decreases dramatically in the knee-flexion posture, with reduced moment arms. To solve this problem, the author recommends using Single Acting Hydraulic Telescopic Cylinder Electro Hydrostatic Actuator (SAT-EHA) to increase the torque of the knee in the knee flexion position. The characteristic of telescopic cylinder is high speed in 1st stage and high force in 2nd stage. The Human Universal Mobility Assist-Hybrid (HUMA-H) was developed by designing and fabricating the waist joint to balance the front and rear directions using an electric motor driver. As the payload capacity increases, the robot pilots can squat and stand up with heavy loads. The performance was verified through the operation test and respiratory gas analysis test of the manufactured HUMA-H.

In this paper, we design and fabricate a wearable walking-assist robot using a tendon-driven method. Most wearable walking-assist robots are designed using the method of the attaching of the motors to the hip, knee, and ankle joints. The robot needs the capacities of the motors attached to the hip and knee joints to equal the weights of the motors attached to the knee and ankle joints and the motor attached to the ankle, respectively. To solve these problems, we design and fabricate the wearable walking-assist robot using a tendon-driven method that rotates the joints by attaching the motors of the hip, knee, and ankle joints to the waist joint, and pulling it with a line. The gait patterns of a normal person are photographed and analyzed, thereby providing the ankle position (x, y) during the walking that is then calculated using the forward kinematic equation, while each joint angle is calculated using the inverse kinematic equation. As a result of the characteristic experiment of the wearable walking-assist robot, the resultant walking aspect is similar to that of the normal person.