Retour au site du LAAS-CNRS

Abstract

Many collision detection methods exist, each specialized for certain data types under certain constraints. In order to enable rapid development of efficient collision detection procedures, we propose an extensible software architecture that allows for cross-queries between data types, while permitting the time and memory optimizations needed for high-performance. By decomposing collision detection into well-defined algorithmic and data components, we can use the same tree-descent algorithm to execute proximity queries, regardless the data type. We validate our implementation on a path planning problem in which a vision guided humanoid represented by an OBB tree explores a dynamic environment composed of voxel maps.

Mots-Clés / Keywords

Abstract

The paper presents a method to control probabilistic diffusion in motion planning algorithms. The principle of the method is to use on line the results of a diffusion algorithm to describe the free space in which the planning takes place. Given that description, it makes the diffusion go faster in favoured directions. That way, if the free space appears as a small volume around a submanifold of a highly dimensioned configuration space, the method overcomes the usual limitations of diffusion algorithms and finds a solution quickly. The presented method is theoretically analyzed and experimentally compared to known motion planning algorithms.

Mots-Clés / Keywords

Abstract

We propose a general and practical planning framework for generating 3-D collision-free motions that take complex robot dynamics into account. The framework consists of two stages that are applied iteratively. In the first stage, a collision-free path is obtained through efficient geometric and kinematic sampling-based motion planning. In the second stage, the path is transformed into dynamically executable robot trajectories by dedicated dynamic motion generators. In the proposed iterative method, those dynamic trajectories are sent back again to the first stage to check for collisions. Depending on the application, temporal or spatial reshaping methods are used to treat detected collisions. Temporal reshaping adjusts the velocity, whereas spatial reshaping deforms the path itself. We demonstrate the effectiveness of the proposed method through examples of a space manipulator with highly nonlinear dynamics and a humanoid robot executing dynamic manipulation and locomotion at the same time.

Mots-Clés / Keywords

Abstract

In this paper, the imitation of human recorded motions by a humanoid robot is considered. The main objective is to reproduce an imitated motion as closely as possible to the original human recored motion. To achieve this goal, the imitation problem is formulated as an optimization problem and the physical limits of the humanoid robot is considered as constraints. The optimization problem is then solved recursively by using an efficient dynamics algorithm, which allows the calculation of the gradient function with respect to the control parameters analytically. The simulation results using OpenHRP platform, which is a dynamical simulator for humanoid robot motions, have pointed out that the imitated motions preserve the salient characteristics of the original human captured motion and the optimization procedure converges well thanks to the analytical calculation of gradient function.

Mots-Clés / Keywords