The Community for Technology Leaders
Green Image
We present an efficient technique that animates human locomotion using balance and comfort control based on inverse dynamics and strength data. Loads or 3D external forces can be applied to any body point. Inverse dynamics using the Newton-Euler method is applied to a 97 DOF human model to compute the joint forces and torques in real time. Balance is maintained by rotating or translating the pelvis and torso. The required torque at each joint is kept below a given fraction of the available torque based on actual human strength data. This "comfort" control adjusts the knee angle or the figure base parameters such as the step length and foot angle. The combination of balance and comfort controls ensures that dynamically sound walking motion is created in each frame. Several visualization techniques are applied to validate and display the result of the dynamics computation, such as the degree of imbalance, the ground reaction force on the foot sole, and the required versus available joint torques. The algorithm also encompasses any walking gait, any figure scale, and any motion path.
animation, simulating humans, inverse dynamics computation, human locomotion

N. I. Badler and H. Ko, "Animating Human Locomotion with Inverse Dynamics," in IEEE Computer Graphics and Applications, vol. 16, no. , pp. 50-59, 1996.
87 ms
(Ver 3.3 (11022016))