Phd Position F/m Anticipating Human Behavior for

Le descriptif de l’offre ci-dessous est en Anglais_

**Type de contrat **:CDD

**Niveau de diplôme exigé **:Bac + 5 ou équivalent

**Fonction **:Doctorant

**A propos du centre ou de la direction fonctionnelle**:
Le centre de recherche Inria de l’Université Grenoble Alpes regroupe un peu moins de 600 personnes réparties au sein de 23 équipes de recherche et 7 services support à la recherche.

Son effectif est distribué sur 3 campus à Grenoble, en lien étroit avec les laboratoires et les établissements de recherche et d'enseignement supérieur (Université Grenoble Alpes, CNRS, CEA, INRAE,), mais aussi avec les acteurs économiques du territoire.

Présent dans les domaines du calcul et grands systèmes distribués, logiciels sûrs et systèmes embarqués, la modélisation de l’environnement à différentes échelles et la science des données et intelligence artificielle, Inria Grenoble - Rhône-Alpes participe au meilleur niveau à la vie scientifique internationale par les résultats obtenus et les collaborations tant en Europe que dans le reste du monde.

**Contexte et atouts du poste**:
**Collaboration between Inria team THOTH and the multidisciplinary institute in artificial intelligence (MIAI) in Grenoble**

The close connection to University of Grenoble Alpes offers additionally many oportunities for collaboration, further training and networking within and cross research fields.

**Mission confiée**:
**Theme/Domain**

**Context and Motivation**
Even after many years of research in human-robot interaction, current systems still struggle to understand users’ intentions, goals, and needs, making it difficult for them to anticipate human actions. The lack of a common representation of behavioral skills that would enable mutual anticipation of actions limits the ability of robots to interact and collaborate effectively with humans. Our goal is to develop a synthetic system capable of context-aware behavior generation that proactively adapts to users’ future actions by monitoring their movements and predicting their intentions during interactions.

**Contact**
The thesis will be jointly supervised by Pia Bideau (THOTH), Karteek Alahari (THOTH) and Xavier Alameda Pineda (RobotLearn).

**Principales activités**:
**Summary**

This project aims to synthesize adaptive behavior, e.g. to adapt the “take” to the “give” in case of human-robot handover tasks. Collaborative tasks, such as human-to-robot handovers, demand a nuanced understanding and anticipation of human actions. These tasks require not only the ability to classify for example the type of the human grasp accurately but also to anticipate future actions effectively.

To achieve this, the project will:

- Identify human hand-object configurations: Develop algorithms to identify and classify the positioning and interaction of human hands with objects. This involves understanding the object's shape and its impact on the human grasp pose.
- Synthesize the corresponding robot grasp: Create robot grasp strategies that are adapted to align with the actions of the human collaborator.

By combining these elements, the project seeks to create a more intuitive but practical human-robot interaction framework. This will enable robots to better understand and predict human behavior, leading to smoother and more intuitive collaborations for a variety of handover tasks.

**Goals**
- Developing adaptable latent action representations: Utilize generative models to create latent action representations that can handle both classification and anticipation tasks. These representations will capture the underlying structure of human actions, allowing the system to recognize and predict behaviors more effectively.
- Addressing challenges of low data availability: Implement strategies to work effectively in environments with limited data. This may involve techniques such as transfer learning, data augmentation, and integrating physical knowledge to ensure robust performance despite scarce training examples.
- Managing initially unknown observations (e.g. unknown action classes): Develop methods to identify and incorporate new, previously unseen classes of actions. This capability will enhance an agent’s ability to function in dynamic and unpredictable environments.

**References**

[1] Halawa, M., Hellwich, O., & Bideau, P. (2022, October). Action-based contrastive learning for trajectory prediction. In _European Conference on Computer Vision_ (ECCV) (pp. 143-159). Cham: Springer Nature Switzerland.

[2] Blume, F., Qu, R., Bideau, P., Maier, M., Rahman, R. A., & Hellwich, O. (2024). How Do You Perceive My Face? Recognizing Facial Expressions in Multi-Modal Context by Modeling Mental Representations. _arXiv preprint arXiv:2409.02566_.

[3] Karamcheti, S., Srivastava, M., Liang, P., & Sadigh, D. (2022, January). Lila: Language-informed latent actions. In _Conference on Robot Learning_ (pp. 1379-1390). PMLR.

[4] Yang, W., Paxton, C., Cakmak, M.,