The project-team E-MOTION aims at developing algorithmic models and methods enabling artificial systems to be built with abilities of perception, decision, and action sufficiently advanced and robust to allow these systems:
- to operate in open and dynamic environments (i.e. in partially known environments, where time and dynamics play a major role),
- and leading to varied interactions with human.
To reach this objective, we propose to combine the respective benefits of the computational geometry, the probability theory and sometimes the biological inspiration (in cooperation with neurophysiologists). These research issues aim at application fields whose purpose is to introduce advanced and secured robotized systems into our "living space" (future cars and transportation systems, or service and intervention robotics).
The research of the project-team E-MOTION is organized along three axes:
- Multimodal and incremental modeling of space and motion (incrementally build several types of models, based on preliminary knowledge and a continuous flow of perceptive data).
- Motion planning for the physical world (simultaneously take into account various constraints of the physical world such as non-collision, environment dynamicity, or reaction time, thanks to techniques developed to reflect on relevant space-time representations).
- Probabilistic inference for decision (correctly reflect on both the current knowledge of the system and its associated uncertainties, by using the new paradigm of bayesian programming developed by our research team).
Keywords:
- Interactive and Cognitive Systems
- Software, Knowledge, Interaction
- Artificial Intelligence, Reasoning and inference, Knowledge, Cognitive models, Human learning
Models, Formal Methods, Mathematics-Informatics interface
- Data discovery and composition, Data querying, Sensor data management , Information retrieval, Data mining
- Decision and action, Interfaces with virtual worlds , Intelligent computers and robotics, Artificial perception, Pattern Recognition
- Programming languages, Model driven engineering, Programming environments
LIG website