S. L. Gay et al.: Identifying and Localizing Dynamic Affordances to Improve Interactions with Other…

Описание:    • 2025 IEEE ICDL Prague   — https://icdl2025.fel.cvut.cz — 2025 IEEE International Conference on Development and Learning (ICDL) Prague ●●
► Simon L. Gay, Olivier Georgeon, Jean-Paul Jamont: Identifying and Localizing Dynamic Affordances to Improve Interactions with Other Agents in Continuous Environments ●●

Allowing autonomous agents to learn by themselves to interact with other agents requires that they be ableto recognize each other and be capable of inferring theirbehaviors. To comply with artificial developmental learning,we follow the radical interactionism hypothesis, in which anagent has no a priori knowledge on its environment. A previouswork has shown that the agent can learn to identify, localize,and predict movements of mobile elements, but was only testedin discrete environments, limiting their applicability on realworld systems. This paper presents new mechanisms for theidentification and localization of mobile entities in a continuousenvironment. These mechanisms learn the relations betweenthe agent’s sensorimotor patterns and the entities, static ormobile, affording them, and store discovered properties in datastructures called Signatures. The properties of signatures arethen exploited to detect distant entities in surrounding environment without relying on a geometrical notion of space. Thesemechanisms were tested in a simple environment, the resultsshowed how signatures integrate the complexity of a continuousenvironment through the limited sensory system of the agent,and localize distant entities through data structures that arecompatible with previously developed behavior inference andprediction mechanisms.