Architectures and Models for Interaction Group

Research topic of the group AMI: Haptic Interaction and Communication

M. Ammi, Y. Bellik, J. Simard, A. Girard, F. Gueniat, Y. Gaffary.

In the landscape of research in haptics, we study and investigate the role of the haptic channel for the analysis and learning of complex and abstract environments and beyond, we would like to investigate the use of the haptic channel for the support of interpersonal communication. We can summarize our research in four main axes.

Characterization of abstract environments. This research begins as a collaboration with the team AERO, dealing with perception of CFD data in the mechanics department at LIMSI. Several haptic rendering algorithms were developed according various constraints of CFD data (discontinuities, variation of gradient, etc.). Afterwards, we proposed to develop a complete approach for the analysis Eulerian structures in complex flow including several vortices. This first experience entailed the Digiteo project (FLUCTUS, Digiteo) aiming at accelerating the calculation of Lagrangian structures. The investigated approach combines new optimization methods and human centered strategies for the orientation of calculation according the progress of calculation.

Collaborative haptics. The role of haptics in molecular structures manipulation was followed by a study on the role of haptics for the coordination of actions of several users during collaborative manipulations of molecular structures. The objective is to improve the distribution of the workload between involved users. This research begins with several experimental studies to understand the contributions and the limits of collaborative strategies for the manipulation of closely coupled structures. These experiments highlight several communication constraints, which limit the working efficiency.  Based on these results, we studied and designed different collaborative metaphors for usual collaborative actions like the collaborative selection, designation of targets (Illustration 4), synchronous manipulation of structures, and collaborative search of targets. These metaphors were studied with new experimental approaches to highlight their roles for the improvement of interpersonal awareness (FRESCO project, ANR).

Affective communication. The use of virtual humans for the communication of emotions showed limitations for the recognition of some emotions, and some difficulties for the differentiation of close emotions. To improve the communication of critical emotions, we propose to complete the visual feedback with a suitable haptic feedback. This research started with the analysis of haptic expressions used by subjects to express a set of emotions. Based on advanced analysis approaches (ANOVA, ACP and EM) according several physical factors (velocity, fluidity, etc.), and perceptual experiments, we defined one haptic expression for each investigated emotion. These haptic expressions were combined with visual expressions to study the complementarity between the two feedbacks, and contribution of the haptic channel to improve the level of recognition and differentiation of close emotions. Beyond the communication of emotions with virtual humans, we investigate and design a haptic interpresence platform for the real time communication of emotions between two remote users (Canal Haptic project, OSEO).

Learning with haptics. This axis includes several collaborations which aim at developing learning platforms for undergraduate students and disabled people. Collaboration with the teams AERO of the mechanics department and CPU of the Human-Machine Communication department at LIMSI has been established in order to design a learning platform for dynamic systems (e.g., pendulum). The proposed approach should improve the understanding of the relation between the real physical system and the corresponding abstract representation (3D attractors) in the phase’s space. A second collaboration with colleagues at IUT of Orsay was initiated to study and design a learning platform for blind people. This platform concerns the study and the exploration of electronic circuits with the haptic channel.

Collaborative Metaphor for Haptic Pointing: Based on the image–plane pointing technique and a spring–time attraction model, we propose to guide physically the partner to the designated target. Experiments were carried out in the context of a molecular deformation procedure and according to different collaborative working strategies. The results show a significant improvement of the performance and efficiency for the different steps of collaborative tasks.