The influence of motion patterns for the best-of-n problem in collective decision-making (MSc)
In this project, we study the influence of motion strategies on collective decisions in mobile multi-agent systems. We will consider 2D environments and can extended it to 3D environments. We study collective decision making with emphasis on the best-of-problems and systems with specific motion patterns that address the needs of the problems at hand.
Collective decision-making is the phenomenon of a collective system making a decision based on the opinions of its individual agents. The agents in such systems are often simple and the decision emerges as a result of their intensive interactions. These interactions are local whereas the consensus that needs to be reached is global. Motion plays a key role in defining these interactions at the agent’s level. In this thesis, the main focus is to explore the influence of different motion patterns on the information propagation within the collective system and aim to find strategies that ensure consensus on the best available option within reasonable time. We consider collective decisions where the system needs to decide between n options, the best-of-n problem emerges where the goal is to have the collective system find the option that best satisfies the system's current needs.
- Design efficient motion strategies in multi-agent systems in 2D and/or 3D
- Analyse the influence of different motion strategies on information spreading
- Establish a fundamental comparison between the developed motion strategies and a set of communication strategies in serving the goal of information spreading in mobile multi-agent system
The work in this master thesis entails:
- Research state-of-the-art communication mechanisms and their current strengths and weaknesses
- Research state-of-the-art on motion strategies in collective systems
- Design motion patterns with the aim of ensuring a well-mixed system that allows each agent to gather information in a uniform fashion, i.e., each agent is maximally informed about all the options within the environment.
- Identify motion strategies that increase the accuracy and/or the speed of the collective decision.
- The impact of interaction models on the coherence of collective decision-making: a case study with simulated locustsY Khaluf, I Rausch, P Simoens, International Conference on Swarm Intelligence, 252-263
- Albert, R., Barabási, A.L.: Statistical mechanics of complex networks. Rev. Mod. Phys. 74, 47–97 (2002)
- Ariel, G., Ayali, A.: Locust collective motion and its modeling. PLOS Comput. Biol. 11(12), 1–25 (2015).
- Memory Induced Aggregation in Collective ForagingJ Nauta, P Simoens, Y Khaluf, International Conference on Swarm Intelligence, 176-189
- Bartumeus, F., da Luz, M.G.E., Viswanathan, G.M., Catalan, J.: Animal search strategies: a quantitative random-walk analysis. Ecology 86(11), 3078–3087 (2005)
- Courses: Programming in Python (INF-22306)
- Required skills/knowledge: Computational modelling, data analytics, interest about collective systems, bio-inspired systems.
Key words: Agent-based modelling, (Agent) Simulation Modelling, Collective systems, decision-making, computational societies
Yara Khaluf (email@example.com)