Retour au site du LAAS-CNRS

Abstract

This paper deals with manipulation task planning for a humanoid robot while stepping. It introduces the concept of documented objects, i.e. objects that provide information on how to manipulate them. The planning phase is decoupled into two parts. First a random motion planner uses the documentation of the object to quickly plan a collision free motion for a simplified model of the robot manipulating the object. Then an inverse kinematics solver animates the whole set of the robot's degrees of freedom by converting the simplified path into time parametrized tasks. Several examples show the generalization of the method.

Abstract

Real trajectories of walking humans can be divided into two major classes: first, nonholonomic trajectories during which the orientation of the human body and the one of the motion are aligned (e.g., straight walking), and second, holonomic ones during which lateral velocities are observed (e.g., side-steps). The major goal of this work is to provide a general locomotion synthesis method for digital actors which enables combining nonholonomic and holonomic walk behaviors. Our motion synthesis technique is based on a motion capture blending method. Such a method is able to transform a lowdimensional trajectory describing the global displacement of a digital actor into a high-dimensional motion which involves all the degrees of freedom of the digital actor body. Our contribution is to extend previously existing motion blending techniques. We consider the desired lateral velocities as a new input of the motion synthesis problem. Thus, the digital actor locomotion is controlled from 3 inputs in our approach: the tangential velocity, the angular velocity and the lateral velocity. The paper describes the major steps of our approach. We first analyze motion captured locomotion cycles to build a Motion Library. The content of the library is projected into the Control Space. Then, at runtime, a desired input is also projected into the Control Space. We deduce a selection of motion captured locomotion cycles from the library, which are finally interpolated to generate a locomotion animation for the digital actors showing the desired input velocities.

Abstract

This paper proposes a unified optimization framework to solve the time-parameterization problem of humanoid-robot paths. Even though the time-parameterization problem is well known in robotics, the application to humanoid robots has not been addressed. This is because of the complexity of the kinematical structure as well as the dynamical motion equation. The main contribution of this paper is to show that the time parameterization of a statically stable path to be transformed into a dynamically stable trajectory within the humanoid-robot capacities can be expressed as an optimization problem. Furthermore, we propose an efficient method to solve the obtained optimization problem. The proposed method has been successfully validated on the humanoid robot HRP-2 by conducting several experiments. These results have revealed the effectiveness and the robustness of the proposed method.

Mots-Clés / Keywords

Mots-Clés / Keywords

Mots-Clés / Keywords