Related Experiment Video
Updated: Jul 7, 2025
06:25
A Real-Time Interactive System for Studying Confrontational Pursuit Behavior in Rodents
Published on: May 16, 2025
159
Translating Virtual Prey-Predator Interaction to Real-World Robotic Environments: Enabling Multimodal Sensing and
Xuelong Sun1,2, Cheng Hu1,3, Tian Liu4
1Machine Life and Intelligence Research Center, Guangzhou University, Guangzhou 510006, China.
Biomimetics (Basel, Switzerland)
|December 22, 2023
Summary
This study bridges virtual and real-world prey-predator dynamics using multi-robot systems with visual and olfactory cues. It reveals emergent behaviors, advancing our understanding of adaptation and embodied intelligence in ecological interactions.
Area of Science:
- Robotics
- Ecology
- Artificial Intelligence
Background:
- Prey-predator interactions are key to evolution and adaptation.
- Agent-based models are popular but struggle with multi-modal interactions.
- Existing models often neglect real-world constraints and noise.
Purpose of the Study:
- To develop a novel prey-predator interaction model.
- To incorporate visual and olfactory sensory cues.
- To bridge the gap between simulations and real-world dynamics.
Main Methods:
- Designed and implemented a prey-predator scenario.
- Utilized both computer simulations and a multi-robot system.
- Incorporated visual and olfactory sensory inputs.
Main Results:
- Observed emergent spatial-temporal dynamics.
- Successfully transitioned prey-predator interaction studies from virtual to real environments.
- Demonstrated the feasibility of multi-robot systems for ecological modeling.
Conclusions:
- Multi-robotics offers a promising approach for studying prey-predator interactions.
- This work lays the foundation for future research on multi-modal sensory processing under real-world constraints.
- Embodied intelligence and emergent behaviors in ecological systems can be investigated through tangible systems.
