Semi-Passive Dynamic Walking Approach for Bipedal Humanoid Robot Based on Dynamic Simulation

Abstract

The research on the principles of legged locomotion is an interdisciplinary endeavor. Such principles are coming together from research in biomechanics, neuroscience, control theory, mechanical design, and artificial intelligence. Such research can help us understand human and animal locomotion in implementing useful legged vehicles. There are three main reasons for exploring the legged locomotion. The first reason is to develop vehicles that can move on uneven and rough terrain. Vehicles with wheels can only move on prepared surfaces such as roads and rails; however, most surfaces are not paved. The second reason is to understand human and animal locomotion mechanics. The study of the mechanisms and principles of control found in nature can help us develop better legged vehicles. The third reason which motivated the study of legged locomotion is the need to build artificial legs for amputees. Although some effective artificial legs have been built to date, more indepth research is required to fully understand the mechanisms and movements necessary to substitute the actual limbs. The research in this paper concerns a group of legged robots known as bipedal walking robots. Research on this subject has a long history; however, it is only in the last two decades that successful experimental prototypes have been developed. The vast majority of humanoid and bipedal robots control the joint angle profiles to carry out the locomotion. Active walking robots (robots with actuators) can do the above task with reasonable speed and position accuracy at the cost of high control efforts, low efficiencies, and most of the time unnatural gaits. WABIAN-2R is among the most successful bipedal walking humanoid robots. In spite of the extensive research on humanoid robots, the actions of walking, running, jumping and manipulation are still difficult for robots. Passive-dynamic walking robots have been developed by researchers to mimic human walking. The main goal of building passive-dynamic walking robots is to study the role of natural dynamics in bipedal walking. Passive-dynamic walkers use gravitational energy to walk down a ramp without any actuators. They are energy efficient but have weak stability in the gait. In addition, the major cause of the energy loss in the current passive-dynamic