More Like this

1. Variable response duration promotes self-organization in decentralized swarms

   https://doi.org/10.1007/978-3-030-63710-1_2  

   Engholdt, Kaelen ; Mathias, H. David ; Wu, Annie S. ( January 2020 , Proceedings of the 9th International Conference on Bioinspired Optimisation Methods and Their Applications)

   null (Ed.)

   In self-organizing multi-agent systems, inter-agent variation is known to improve swarm performance significantly. Response duration, the amount of time that an agent spends on a task, has been proposed as a form of inter-agent variation that may be beneficial. In the biological literature, variability in agent response duration in natural swarms for desynchronizing agent actions has been discussed for some time. This form of variation, however, is not well understood in artificial swarms. In this work, we explore inter-agent variation in response duration as a desynchronization technique. We find that variation in response duration does desynchronize agent behaviors and does improve swarm performance on a two-dimensional tracking problem in which the swarm must push a tracker, staying as close as possible to a moving target. By preventing agents from reacting identically to task stimuli and keeping some agents on task longer, response duration helps smooth the swarm’s path and allows it to better track the target into path features such as corners. 

   more » « less
2. Dynamic response thresholds: heterogeneous ranges allow specialization while mitigating convergence to sink states

   Wu, Annie S. ; Mathias, H. David ( January 2020 , Proceedings of the 12th International Conference on Swarm Intelligence)

   null (Ed.)

   We argue that heterogeneous threshold ranges allow agents in a decentralized swarm to effectively adapt thresholds in response to dynamic task demands while avoiding the pitfalls of positive feedback sinks. Dynamic response thresholds allow agents to dynamically evolve specializations which can improve the responsiveness and stability of a swarm. Dynamic thresholds that adapt in response to previous experience, however, are vulnerable to getting stuck in sink states due to the positive feedback nature of such systems. We show that heterogeneous threshold ranges result in comparable task allocation and improved stability as compared to homogeneous threshold ranges, and that simple static random thresholds should be considered in situations where agent resources are plentiful. 

   more » « less
3. Decentralized swarm desynchronization via inter-agent variation for logistic resupply

   Giordano, Joseph P. ; Wu, Annie S. ; Pherwani, Arjun ; Mathias, H. David ( January 2021 , Proceedings of the 34th Florida Artificial Intelligence Research Society Conference)

   null (Ed.)

   Decentralized computational swarms have been used to simulate the workings of insect colonies or hives, often utilizing a response threshold model which underlies agent interaction with dynamic environmental stimuli. Here, we propose a logistics resupply problem in which agents must select from multiple incoming scheduled tasks that generate competing resource demands for workers. This work diverges from previous attempts toward analyzing swarm behaviors by examining relative amounts of stress placed on a multi-agent system in conjunction with two mechanisms of response: variable threshold distribution, or duration level. Further, we demonstrate changes to the general swarm performance’s dependence on paired desynchronization type and schedule design, as the result of varied swarm conditions. 

   more » « less
4. Response thresholds generalize across problem instances for a deterministic response multiagent system

   Mathias, H. David ; Wu, Annie S. ; and Dang, Daniel ( January 2021 , Proceedings of the Genetic and Evolutionary Computation Conference)

   null (Ed.)

   In this work, we use a multiobjective genetic algorithm to evolve agent response thresholds for a decentralized swarm and demonstrate that swarms with evolved thresholds outperform swarms with thresholds set using other methods. In addition, we provide evidence that the effectiveness of evolved thresholds is due in part to the evolutionary process being able to find, not just good distributions of thresholds for a given task across all agents, but also good combinations of thresholds over all tasks for individual agents. Finally, we show that thresholds evolved for some problem instances can effectively generalize to other problem instances with very different task demands. 

   more » « less
5. Dirichlet-Multinomial Counterfactual Rewards for Heterogeneous Multiagent Systems

   Dixit, G. ; Zerbel, N ; Tumer, K. ( January 2019 , IEEE Symposium on multi-robot and multi-agent systems)

   Multi-robot teams have been shown to be effective in accomplishing complex tasks which require tight coordination among team members. In homogeneous systems, recent work has demonstrated that “stepping stone” rewards are an effective way to provide agents with feedback on potentially valuable actions even when the agent-to-agent coupling require- ments of an objective are not satisfied. In this work, we propose a new mechanism for inferring hypothetical partners in tightly-coupled, heterogeneous systems called Dirichlet-Multinomial Counterfactual Selection (DMCS). Using DMCS, we show that agents can learn to infer appropriate counterfactual partners to receive more informative stepping stone rewards by testing in a modified multi-rover exploration problem. We also show that DMCS outperforms a random partner selection baseline by over 40%, and we demonstrate how domain knowledge can be used to induce a prior to guide the agent learning process. Finally, we show that DMCS maintains superior performance for up to 15 distinct rover types compared to the performance of the baseline which degrades rapidly. 

   more » « less