Parallel Inverse Kinematics for Multithreaded Architectures

Abstract

In this article, we present a parallel prioritized Jacobian-based inverse kinematics algorithm for multithreaded architectures. We solve damped least squares inverse kinematics using a parallel line search by identifying and sampling critical input parameters. Parallel competing execution paths are spawned for each parameter in order to select the optimum that minimizes the error criteria. Our algorithm is highly scalable and can handle complex articulated bodies at interactive frame rates. We show results on complex skeletons consisting of more than 600 degrees of freedom while being controlled using multiple end effectors. We implement the algorithm both on multicore and GPU architectures and demonstrate how the GPU can further exploit fine-grain parallelism not directly available on a multicore processor. Our implementations are 10 to 150 times faster compared to a state-of-the-art serial implementation while providing higher accuracy. We also demonstrate the scalability of the algorithm over multiple scenarios and explore the GPU implementation in detail.

Inverse Kinematics Techniques in Computer Graphics: A Survey

“…Nevertheless, some effort has been recently devoted to parallel computation of the Jacobian based methods, e.g. [HMCB16], to gain efficiency and deal with singularities and multiple path problems.…”

“…Similarly, Harish et al . [HMCB16] extend the DLS method to exploit maximum parallelism by mapping the internal DLS steps to the data‐parallel GPU architecture. Huang et al .…”

“…More specifically, Farzan and DeSouza [FD13] proposed a parallelization method that relies on the computation of multiple numerical estimations of the Jacobian inverse; they handle singularities and multiple paths in redundant robots by selecting the best path to the desired configuration of the end effector on multi-core architectures for the Denavit-Hartenberg representation of robots. Similarly, Harish et al [HMCB16] extend the DLS method to exploit maximum parallelism by mapping the internal DLS steps to the dataparallel GPU architecture. Huang et al [HFDP16] employ a variation of the SDLS method, while the joint motion parameters are learned automatically from pre-captured motion data that are stored in an octree for fast access.…”

Inverse kinematics using dynamic joint parameters: inverse kinematics animation synthesis learnt from sub-divided motion micro-segments

“…Harish et al [14] propose a parallel IK architecture for high degree of freedom models. They also use the Damped Least Squares Inverse Kinematics method.…”

Interactive character posing with efficient collision handling

“…In this respect, the IK is especially useful when editing the pose of a character that has many degrees of freedom in the hierarchical skeleton structure. Thanks to long‐standing studies for IK for several decades, its speed, stability, and precision have been drastically improved, and recent IK frameworks allow the user to manipulate the character pose interactively.…”