Trot gait based feed-forward walking on challenging terrain: Case of high step climbing

For a quadruped robot, the crawl gait is commonly used on rough terrain environment where a pre-motion planing is required. However its walking speed is slower than other walking gaits and make it difficult to develop a practical legged robot. Among the various types of gaits, we consider that the trot gait is safety gait which can prevent complete falling by hitting its swinging leg on the ground. Therefore in this paper we propose a new feed-forward trot gait algorithm based on 3D sway compensation trajectory which can control footsteps and position of the center of gravity for rough terrain environment. The validity of the proposed algorithm is confirmed by step climbing experiment using a spawling-type quadruped robot TITAN-XIII. As a result, the robot climbed up the step of 120 mm height which is 40% of its height in 3 seconds and proved the feasibility of the trot gait on a difficult terrain.