Bruno Arnaldi

A new space subdivision method for ray tracing CSG modelled scenes

A new algorithm for space tracing with CSG modelled scenes is presented. Space is divided in an irregular fashion to fit the objects as closely as possible. For this reason, primitive minimal bounding boxes are computed. Space subdivision is achieved in two steps: partitioning in projection plane and depth partitioning. A set of 3D regions named cells are then created. A Boolean CSG tree is distributed into the cell structure to form in each cell the minimal boolean CSG tree using the relevant primitives. The searching process for the “next cell” along the ray path is performed by using a local data structure associated with each cell. The goal of this structure is to link the cells together. An improvement, named “mailbox”, for all space tracing algorithms is detailed. Results are presented for two scenes to compare this new algorithm with Roths algorithm.

STORM: a Generic Interaction and Behavioral Model for 3D Objects and Humanoids in a Virtual Environment

We present in this paper research focused on interaction and behavior specification for 3D objects in general, including humanoids. This research is conducted in the context of a collaboration with Nexter-Group (a French military manufacturer) in order to introduce Virtual Reality (VR) in maintenance training. The use of VR environments for training is strongly stimulated by important needs of training on sensitive equipment, sometimes fragile, unavailable, costly or dangerous. Nevertheless, for the development of such applications, the re-use of existing developments is a major issue. Our research is focused on models that have been designed to achieve this reusability and standardization for the efficient development of new virtual environments. In particular, we defined a new generic model named STORM, used to describe reusable behaviors for 3D objects in general, including humanoids, and reusable interactions between those objects.

A software system to carry-out virtual experiments on human motion

This work presents a simulation system designed to carry-out virtual experiments on human motion. 3D visualization, automatic code generation and generic control design patterns provide biomechanicians and medics with dynamic simulation tools. The paper first deals with the design of mechanical models of human beings. It also presents design patterns of controllers for an upper-limb model composed of 11 degrees of freedom. As an example, two controllers are presented in order to illustrate these design patterns. The paper also presents a user-friendly interface dedicated to medics that makes it possible to enter orders in natural language.

Dynamics and Unification of Animation Control

We present an analysis of unified animation control processes. The design of an animation system is strongly coupled with the considered application. Most of the existing systems are specialized. For example, in audiovisual production or in C.A.D., only a few systems specialized in mechanical computation have capabilities in the image synthesis field. We are working on the design of a general animation system, including generation of motion defined by use of mechanical laws, in an audiovisual environment. Our system is built around a structured graph in which we store a hierarchical description of the objects and mechanical joints. The joints may be used to link the objects together to allow the building of multibody mechanical systems. For objects which are not submited to mechanical laws, motion control can be specified by key-framing techniques or explicit trajectories. We use a dynamical formalism based upon the principle of virtual work associated with LAGRANGEs multipliers and that takes into account holonomic and nonholonomic constraints. The generation of the differential motion equations system is automatically built by performing symbolic derivations. These equations are then solved for each time step and give object locations and orientations.

Coarse-to-fine design of feedback controllers for dynamic locomotion

Control engineers have long known how to design efficient systems for the guidance and stabilization of man-made machines such as aircraft, submarines, and automatic cars. In order to make synthetic animals more closely aresembleo the systems for which control engineers already possess some design solutions, the propelling muscles of the animal are first initiated by the continuous motors found in man-made machinery. A linear feedback controller can then be designed at this coarse level. Finally, the model is refined and a neuromuscular subsystem is added to mimic the pulsed propulsion of a real animal.

Human motion coordination: example of a juggler

The paper introduces a novel method for the coordination of human motion based on planification and AI techniques. Motions are considered as black boxes that are activated according to pre-conditions and produce post-conditions in a hybrid continuous and discrete world. Each part of the body is an autonomous entity which cooperates with the others depending on global criteria such as occupation rate and distance to a goal (common to all the entities). This technique makes it possible to easily specify and solve the motion coordination problem of a juggler that deals with a dynamic number of balls in real time.

Pulse-Modulated Locomotion for Computer Animation

In this paper, we address the problem of animating and controlling the locomotion of bodies which are propelled by pulse-like and periodic muscle activation. This kind of locomotion mode is encountered in a variety of real animals and an even larger variety of imaginary creatures could be propelled in this way. We propose a step-by-step methodology for the design of feedback controllers that can propel a dynamic model with an “intelligent” pulse-train of muscle efforts. The idea is to proceed in three steps: (i) a feedback controller is designed just as if muscles could deliver continuously regulated efforts. (ii) a pulse-modulator is inserted into the feedback controlled system, so as to take the muscle activation pattern into account. (iii) Synchronised kinematic cosmetics are added to the motion. The technique is illustrated by applying it to dynamic models of increasing complexity.

Quick tuning of a reference locomotion gait [computer animation]

This paper presents a comprehensive system for adaptive replay of a computer animation sequence featuring the locomotion of a synthetic animal. The kinematic locomotion gait, which is supposed to be mostly stationary, is tuned before being injected on the dynamic model of the animated animal. The animator can take benefit from this new flexibility in order to fulfil some goals, like steering the animal towards a target. The reference gait kinematic trajectories are expanded on a wavelet redundant bases, and tuned through varying the wavelet parameters. Because the optimization process is time-consuming, we propose to carry-out all optimizations beforehand and send the outcomes in a training set for a neural network, whose role is to compress data and to interpolate in between the training samples. At runtime, the network emulates the behaviour of the optimization process and automatically tailors the gait parameters. This approach paves the way towards the definition of a ready-to-use model for animated gaits.<<ETX>>

VISYR: a simulation tool for mobile robots using vision sensors

With the advent of mobile robots and inboard vision sensors mounted directly on the robots wrist, new kinds of problems lie in the image processing field as, for example, dynamic scene analysis or motion estimation. The lack of flexibility of real experiments led us to implement at IRISA a general simulation tool devoted to the study of robots using moving vision sensors. VISYR allows us to simulate the image perceived by a robot of its environment during its motion.The first part of the paper is devoted to the modelling of the 3D scene containing complex objects and to the design of a suitable robotics vision sensor. In the second part, a new algorithm of dynamic management of the local data basis perceived by the sensor is presented. The parameters of the vision sensors are highly adjustable and VISYR is conceived to allow the fast development of algorithms using dynamic vision data.