Abstract
This paper describes the experimental realization of dynamic walking by a six-legged robot, Little Crabster, on uneven terrain. Dynamic walking is achieved through the processes of walking pattern generation and posture stabilization. A wave gait that sequentially moves the legs with the greatest degree of walking stability is chosen as the walking pattern, and predesigned gait trajectories are generated according to the proposed walking parameters. In addition, the pattern is modified online through the use of ground contact information from the six feet. Posture stabilization consists of CoP (Center of Pressure) control to maintain a dynamic balance against external disturbances, body posture control to maintain the level body, and landing control to adapt to uneven ground with a small landing impact. These controls are addressed in detail. Finally, the performance of the proposed six-legged walking algorithm is experimentally verified through a walking experiment on a large treadmill with a global slope and obstacles on the floor.
Original language | English |
---|---|
Pages (from-to) | 509-518 |
Number of pages | 10 |
Journal | International Journal of Precision Engineering and Manufacturing |
Volume | 18 |
Issue number | 4 |
DOIs | |
State | Published - 1 Apr 2017 |
Keywords
- CoP control
- Six-legged robot
- Uneven terrain
- Walking pattern