Emerging Inter-Swarm Collaboration for Surveillance Using Pheromones and Evolutionary Techniques
Daniel H Stolfi1, Matthias R Brust1, Grégoire Danoy1,2
1SnT, University of Luxembourg, L-4364 Esch-sur-Alzette, Luxembourg.
Sensors (Basel, Switzerland)
|May 6, 2020
Summary
A new mobility model, Attractor Based Inter-Swarm collaborationS (ABISS), enhances surveillance using diverse autonomous vehicles. This collaborative approach improves coverage in restricted areas by up to 11%.
Area of Science:
- Robotics and Autonomous Systems
- Artificial Intelligence
- Operations Research
Background:
- Effective surveillance of restricted areas is crucial.
- Current autonomous vehicle (AV) strategies often lack inter-swarm collaboration for complex terrains.
- Unpredictable trajectories are needed for thorough exploration.
Purpose of the Study:
- To introduce the Attractor Based Inter-Swarm collaborationS (ABISS) mobility model.
- To enhance the surveillance capabilities of unmanned autonomous vehicles in restricted areas.
- To demonstrate the effectiveness of inter-swarm collaboration.
Main Methods:
- Utilizing diverse vehicle types (e.g., ground, aerial) with chaotic trajectories.
- Implementing an evolutionary algorithm for parameterizing and configuring collaborative strategies.
- Simulating and comparing the ABISS model against non-collaborative approaches.
Main Results:
- Collaboration between different vehicle swarms is demonstrated as feasible.
- The ABISS model achieved up to an 11% improvement in total covered area.
- Collaborative configurations emerged naturally from the evolutionary algorithm.
Conclusions:
- The ABISS model significantly improves surveillance coverage through inter-swarm collaboration.
- Evolutionary algorithms are effective in optimizing collaborative autonomous vehicle strategies.
- ABISS offers a promising solution for complex area surveillance missions.
