Using task efficient contact configurations to animate creatures in arbitrary environments

A common issue in three-dimensional animation is the creation of contacts between a virtual creature and the environment. Contacts allow force exertion, which produces motion. This paper addresses the problem of computing contact configurations allowing to perform motion tasks such as getting up from a sofa, pushing an object or climbing. We propose a two-step method to generate contact configurations suitable for such tasks. The first step is an offline sampling of the range of motion (ROM) of a virtual creature. The ROM of the human arms and legs is precisely determined experimentally. The second step is a run time request confronting the samples with the current environment. The best contact configurations are then selected according to a heuristic for task efficiency. The heuristic is inspired by the force transmission ratio. Given a contact configuration, it measures the potential force that can be exerted in a given direction. The contact configurations are then used as inputs for an inverse kinematics solver that will compute the final animation. Our method is automatic and does not require examples or motion capture data. It is suitable for real time applications and applies to arbitrary creatures in arbitrary environments. Various scenarios (such as climbing, crawling, getting up, pushing or pulling objects) are used to demonstrate that our method enhances motion autonomy and interactivity in constrained environments. & 2014 Elsevier Ltd. All rights reserved. Research in computer animation is motivated by the need to provide virtual creatures with an increased autonomy of motion in 3D environments. Such improvements allow to propose new forms of gameplay in video games, or to validate ergonomic designs. In this work we are interested in the contacts created between a creature and the environment: contacts allow to efficiently exert the force necessary to perform motion tasks (such as getting up, climbing or pulling). For instance in Fig. 13, several contacts are created between the end-effectors of a virtual insect and the books composing the environment. Motion capture methods are inherently limited in such a constrained context: addressing various tasks and environments for different creatures requires the creation of prohibitively large motion databases. Therefore, a common approach is the decomposition of the motion into a sequence of contact configurations between a virtual creature and the environment. The notion of configuration is central in motion planning [1]. Such planners often use randomly generated configurations [2], and select those preserving static stability [3]. However, they lack heuristics to determine if those configurations are suited for the task in terms of force exertion. In the rest of the paper such configurations are called task efficient. Dynamic simulations use predefined configurations as inputs to motion controllers, but show little adaptation to the environment [4]. Thus, motion planners and dynamic controllers could benefit from a method to generate appropriate contact configurations. This is our problem statement, formalized in Section 2. The key idea: The environment as a mean to exert a force: Contacts allow force exertion, which in turn produces the motion. Therefore to select a contact configuration, it is important to make sure it will allow to perform the task. For this reason we need heuristics to measure the compatibility of a contact configuration with a translational motion task. Examples of such tasks are pushing, pulling, standing up, or climbing. This set of motions is commonly needed by interactive simulations (such as videogames). They could benefit from our method to introduce more variety in the environments and interactions they propose. Rotational tasks will be addressed in future works. To measure the task efficiency, we propose a heuristic inspired by the force transmission ratio [5]. It defines the efficiency of a configuration as the potential force it allows to exert in the direction of a translational task, as detailed in Section 2.4. It is traditionally used to optimize the configuration of a robotic arm, but requires to