We present a unified dynamics framework for real-time visual effects. Using particles connected by constraints as our fundamental building block allows us to treat contact and collisions in manner, we show how this representation is flexible enough model gases, liquids, deformable solids, rigid bodies cloth with two-way interactions. address some common problems traditional particle-based methods describe parallel constraint solver based on position-based that efficient applications.

We address the long-standing problem of iteration count and time step dependent constraint stiffness in position-based dynamics (PBD). introduce a simple extension to PBD that allows it accurately efficiently simulate arbitrary elastic dissipative energy potentials an implicit manner. In addition, our method provides force estimates, making applicable wider range applications, such those requiring haptic user-feedback. compare algorithm more expensive non-linear solvers find produces...

This paper presents a method for animating fluid using unstructured tetrahedral meshes that change at each time step. We show conform well to changing boundaries and focus computation in the visually important parts of domain can be generated quickly reliably existing techniques. also describe new approach two-way coupling rigid bodies that, while general, benefits from remeshing. Overall, provides flexible environment creating complex scenes involving animation.

We present algorithms for simulating and visualizing the insertion steering of needles through deformable tissues surgical training planning. Needle is an essential component many clinical procedures such as biopsies, injections, neurosurgery, brachytherapy cancer treatment. The success these depends on accurate guidance needle tip to a target while avoiding vital tissues. deforms body tissues, making placement difficult. Our interactive simulator models coupling between steerable tissue....

We present a new Eulerian fluid simulation method, which allows real-time simulations of large scale three dimensional liquids. Such scenarios have hitherto been restricted to the domain off-line computation. To reduce computation time we use hybrid grid representation composed regular cubic cells on top layer tall cells. With this layout water above an arbitrary terrain can be represented without consuming excessive amount memory and compute power, while focusing effort area near surface...

Most Deep Reinforcement Learning (Deep RL) algorithms require a prohibitively large number of training samples for learning complex tasks. Many recent works on speeding up RL have focused distributed and simulation. While is often done the GPU, simulation not. In this work, we propose using GPU-accelerated simulations as an alternative to CPU ones. Using NVIDIA Flex, GPU-based physics engine, show promising speed-ups various continuous-control, locomotion With one GPU core, are able train...

We present a new Eulerian fluid simulation method, which allows real-time simulations of large scale three dimensional liquids. Such scenarios have hitherto been restricted to the domain off-line computation. To reduce computation time we use hybrid grid representation composed regular cubic cells on top layer tall cells. With this layout water above an arbitrary terrain can be represented without consuming excessive amount memory and compute power, while focusing effort area near surface...

We propose a new method for both collision detection and response geared towards handling complex deformable objects in close contact. Our does not miss events between time steps solves the challenging problem of untangling automatically robustly. It is conceptually simple straight forward to parallelize due regularity algorithm. The main idea tessellate air once before simulation by considering one unilateral constraint per element that prevents its inversion during simulation. If large...

We introduce a deep reinforcement learning method that learns to control articulated humanoid bodies imitate given target motions closely when simulated in physics simulator. The motion, which may not have been seen by the agent and can be noisy, is supplied at runtime. Our recover balance from moderate external disturbances keep imitating motion. When subjected large cause fall down, our character get up track trained mocap clips CMU motion capture database number of other publicly...

We describe a method for animating incompressible liquids with detailed free surfaces. For each time step, semi-Lagrangian contouring computes new fluid boundary (represented as fine surface triangulation) from the previous step's and velocity field. Then mesh generation algorithm called isosurface stuffing discretizes region enclosed by boundary, creating tetrahedral that grades resolution at to coarser in interior. The has structure, based on body centered cubic lattice, accommodates...

We propose a new fast and robust method to simulate various types of solid including rigid, plastic soft bodies as well one, two three dimensional structures such ropes, cloth volumetric objects. The underlying idea is use oriented particles that store rotation spin, along with the usual linear attributes, i.e. position velocity. This additional information adds substantially traditional particle methods. First, can be represented by anisotropic shapes ellipsoids, which approximate surfaces...

We present a hybrid water simulation method that combines grid based and particles approaches. Our specialized shallow solver can handle arbitrary underlying terrain slopes, depth supports wet-dry regions tracking. To treat open scenes we introduce for handling non-reflecting boundary conditions. Regions of liquid cannot be represented by the height field including breaking waves, falls splashing due to rigid soft bodies interaction are automatically turned into spray, splash foam particles....

We propose a new fast and robust method to simulate various types of solid including rigid, plastic soft bodies as well one, two three dimensional structures such ropes, cloth volumetric objects. The underlying idea is use oriented particles that store rotation spin, along with the usual linear attributes, i.e. position velocity. This additional information adds substantially traditional particle methods. First, can be represented by anisotropic shapes ellipsoids, which approximate surfaces...

We propose a new fast, robust and controllable method to simulate the dynamic destruction of large complex objects in real time. The common for fracture simulation computer games is pre-fracture models replace by their pre-computed parts at run-time. This popular computationally cheap but has disadvantages that pattern does not align with impact location number hierarchical levels fixed. Our allows fracturing into an unlimited pieces fast enough be used games. represent visual meshes...

Signed distance fields (SDFs) are a popular shape representation for collision detection. This is due to their query efficiency, and the ability provide robust inside/outside information. Although it straightforward test points interpenetration with an SDF, not clear how extend this continuous surfaces, such as triangle meshes. In paper, we propose per-element local optimization find closest between SDF isosurface mesh elements. allows us generate accurate contact sharp point-face pairs,...

Bevel-tip steerable needles are a promising new technology for improving accuracy and accessibility in minimally invasive medical procedures.As yet, 3D needle steering has not been demonstrated the presence of tissue deformation uncertainty, despite application progressively more sophisticated planning algorithms.This paper presents feedback controller that steers along helical paths, varies helix radius to correct perturbations.It achieves high targets sufficiently far from insertion point;...

We present algorithms for simulating and visualizing the insertion steering of needles through deformable tissues surgical training planning. Needle is an essential component many clinical procedures such as biopsies, injections, neurosurgery, brachytherapy cancer treatment. The success these depends on accurate guidance needle tip to a target while avoiding vital tissues. deforms body tissues, making placement difficult. Our interactive simulator models coupling between steerable tissue....

We present a simple and fast method to add wrinkles dynamic meshes such as simulated cloth or the skin of an animated character. To get desired surface details, we attach higher resolution wrinkle mesh coarse base allowing vertices deviate from their attachment positions within limited range. The shape is determined by static solver which runs in parallel motion mesh. Our can be used automatically enhance purely with would tedious task do hand. fact that tessellation chosen independently...

In this paper we re-examine the idea that implicit integrators with large time steps offer best stability/performance trade-off for stiff systems. We make surprising observation performing a single step n constraint solver iterations is less effective than computing smaller steps, each iteration. Based on observation, our approach to split every visual into substeps of length Δt/n and perform iteration extended position-based dynamics (XPBD) in such substep. When compared traditional...

Abstract We present a rigid body simulation method that can resolve small temporal and spatial details by using quasi explicit integration scheme is unconditionally stable. Traditional simulators linearize constraints because they operate on the velocity level or solve equations of motion implicitly thereby freezing constraint directions for multiple iterations. Our always works with most recent directions. This allows us to trace high speed objects colliding against curved geometry, reduce...

This paper presents a method for simulating the two-way interaction between fluids and deformable solids. The are simulated using an incompressible Eulerian formulation where linear pressure projection on fluid velocities enforces mass conservation. Similarly, elastic solids semi-implicit integrator implemented as operator applied to forces acting nodes in Lagrangian formulation. proposed coupling constraints systems by combining both implicit integration steps into one set of simultaneous...

AbstractIn this paper we focus on the fast simulation of hair and fur animated characters. While it is common in featurefilms to simulate computer generated actors, characters are still mostly hand-animated computergames. A main difficulty simulating that perceived as inextensible by humans. Preventing an objectfrom being stretched a global, non-linear problem. This reason why completely inextensibleobjects real-time remains major challenge open research topic.Existing approaches typically use...

We present a framework for the simulation of rigid and deformable bodies in presence contact friction. Our method is based on non-smooth Newton iteration that solves underlying nonlinear complementarity problems (NCPs) directly. This approach allows us to support dynamics models, including hyperelastic articulated mechanisms, coupled through smooth isotropic friction model. The fixed-point nature our means it requires only solution symmetric linear system as building block. propose new...