Displacement constraints for interactive modeling and animation of articulated structures

Abstract

PublishedInternational audienceThis paper presents an integrated set of methods for the automatic construction and interactive animation of solid systems that satisfy specified geometric constraints. Displacement contraints enable the user to design articulated bodies with various degrees of freedom in rotation or in translation at highes and to restrict the scope of the movement at will. The graph of constrained objects may contain closed loops. The animation is achieved by decoupling the free motion of each solid component from the action of the constraints. We do this with iterative tunings in displacements. The method is currently implemented in a dynamically based animation system and takes the physical parameters into account while reestablishing the constraints. In particular, first-order momenta are preserved during this process. The approach would be easy to extend to modeling systems or animation modules without a physical model just by allowing the user to control more parameters. (source Springer

Unified particle physics for real-time applications

“…Rather, we prefer to combine custom position-level constraints for each material we wish to simulate. Gascuel and Gascuel [1994] used displacement constraints to animate rigid bodies and articulated figures. Their work was extended by Faure [1999], and position-based dynamics [Müller et al 2007] can be seen as a generalization of their method.…”

Interactive Solid Animation Using Linearized Displacement Constraints

“…Runge-Kutta, and denote the result q t + dt ;ṽ t + dt perform post-stabilizationin order to meet the geometric constraints and their first derivatives: q t + dt =q t + dt , q (5) v t + dt =ṽ t + dt , v (6) The computation of the correction terms is explained in section 3.3. This approach frees the user from tuning arbitrary stabilization parameters, and experiments have shown better stability than Baumgarte stabilization [4].…”

“…Compared with previous related work [5], the efficiency of our method comes from the use of the biconjugate gradient algorithm, along with a robust integration scheme.…”

Dynamic simulation of articulated rigid bodies with contact and collision

“…These techniques project the solution back onto the acceptable constraint manifold after each time step. Ad hoc methods for correcting the positions and orientations were proposed in [25] (see also the procedural approach in [26]) correcting the velocities based on the final state, and an improved iterative method was proposed in [27] (see also [8]). Our approach has notable differences such as solving a set of nonlinear equations locally and using impulses (as opposed to displacements) so that the approach can be combined with state of the art contact and collision algorithms.…”