Animating human athletics

Abstract

This paper describes algorithms for the animation of men and women performing three dynamic athletic behaviors: running, bicycling, and vaulting. We animate these behaviors using control algorithms that cause a physically realistic model to perform the desired maneuver. For example, control algorithms allow the simulated humans to maintain balance while moving their arms, to run or bicycle at a variety of speeds, and to perform a handspring vault. Algorithms for group behaviors allow a number of simulated bicyclists to ride as a group while avoiding simple patterns of obstacles. We add secondary motion to the animations with springmass simulations of clothing driven by the rigid-body motion of the simulated human. For each simulation, we compare the computed motion to that of humans performing similar maneuvers both qualitatively through the comparison of real and simulated video images and quantitatively through the comparison of simulated and biomechanical data.

Simulating balance recovery responses to trips based on biomechanical principles

“…Therefore, we employ finite state machines (FSM) to model the responses. This FSM approach has also been used in computer graphics to construct controllers for running and other behaviors [RH91, HWBO95,YLvdP07]. The control laws for each state of the FSM are modeled from the motion capture data.…”

“…Their idea was to design a finite state machine based on contact states between feet and ground and then design control actions for each joint in each state. This idea was extended to a running, vaulting, and diving human character by Hodgins and colleagues [HWBO95]. Yin and colleagues implemented FSM-based controllers for various behaviors such as a back flip and a skip [YLvdP07].…”

Fluid Simulation with Articulated Bodies

“…Methods for defining controllers for complex behaviors [28] to learning simple controllers [29] have been proposed. However, most of these methods are more suitable for forward dynamic simulation [30] improving on efforts in dynamic robotic manipulation [31].…”

Pose Control in Dynamic Conditions

“…Animation systems can arrange the possible poses in a directed graph, with edges representing allowed transitions and runtime execution traversing this graph [7] according to user input [17] or the game's artificial intelligence (AI). Each pose specifies target positions for each controlled degree of freedom (DoF).…”

“…At run-time, the character's joints apply torques to reach the target pose, with the specific amount of torque supplied computed by a low-level control law, such as the proportional-derivative (PD). Despite (or perhaps, because of) the simplicity of pose control, it is commonly used in practice and in the animation literature, where it often forms the foundation of more complex algorithms capable of impressive behaviors [7,4,12,16,11,15].…”