We propose Heterogeneous Swarms, an algorithm to design multi-LLM systems by jointly optimizing model roles and weights. We represent multi-LLM systems as directed acyclic graphs (DAGs) of LLMs with topological message passing for collaborative generation. Given a pool of LLM experts and a utility function, Heterogeneous Swarms employs two iterative steps: role-step and weight-step. For role-step, we interpret model roles as learning a DAG that specifies the flow of inputs and outputs between LLMs. Starting from a swarm of random continuous adjacency matrices, we decode them into discrete DAGs, call the LLMs in topological order, evaluate on the utility function (e.g.

We study the problem of determining the emergent behaviors that are possible given a functionally heterogeneous swarm of robots with limited capabilities. Prior work has considered behavior search for homogeneous swarms and proposed the use of novelty search over either a hand-specified or learned behavior space followed by clustering to return a taxonomy of emergent behaviors to the user. In this paper, we seek to better understand the role of novelty search and the efficacy of using clustering to discover novel emergent behaviors. Through a large set of experiments and ablations, we analyze the effect of representations, evolutionary search, and various clustering methods in the search for novel behaviors in a heterogeneous swarm. Our results indicate that prior methods fail to discover many interesting behaviors and that an iterative human-in-the-loop discovery process discovers more behaviors than random search, swarm chemistry, and automated behavior discovery. The combined discoveries of our experiments uncover 23 emergent behaviors, 18 of which are novel discoveries. To the best of our knowledge, these are the first known emergent behaviors for heterogeneous swarms of computation-free agents. Videos, code, and appendix are available at the project website: https://sites.google.com/view/heterogeneous-bd-methods

There has been growing interest in the use of multi-robot systems in various tasks and scenarios. The main attractiveness of such systems is their flexibility, robustness, and scalability. An often overlooked yet promising feature is system modularity, which offers the possibility to harness agent specialization, while also enabling system-level upgrades. However, altering the agents' capacities can change the exploration-exploitation balance required to maximize the system's performance. Here, we study the effect of a swarm's heterogeneity on its exploration-exploitation balance while tracking multiple fast-moving evasive targets under the Cooperative Multi-Robot Observation of Multiple Moving Targets framework. To this end, we use a decentralized search and tracking strategy with adjustable levels of exploration and exploitation. By indirectly tuning the balance, we first confirm the presence of an optimal balance between these two key competing actions. Next, by substituting slower moving agents with faster ones, we show that the system exhibits a performance improvement without any modifications to the original strategy. In addition, owing to the additional amount of exploitation carried out by the faster agents, we demonstrate that a heterogeneous system's performance can be further improved by reducing an agent's level of connectivity, to favor the conduct of exploratory actions. Furthermore, in studying the influence of the density of swarming agents, we show that the addition of faster agents can counterbalance a reduction in the overall number of agents while maintaining the level of tracking performance. Finally, we explore the challenges of using differentiated strategies to take advantage of the heterogeneous nature of the swarm.

Current strategies employed for maritime target search and tracking are primarily based on the use of agents following a predetermined path to perform a systematic sweep of a search area. Recently, dynamic Particle Swarm Optimization (PSO) algorithms have been used together with swarming multi-robot systems (MRS), giving search and tracking solutions the added properties of robustness, scalability, and flexibility. Swarming MRS also give the end-user the opportunity to incrementally upgrade the robotic system, inevitably leading to the use of heterogeneous swarming MRS. However, such systems have not been well studied and incorporating upgraded agents into a swarm may result in degraded mission performances. In this paper, we propose a PSO-based strategy using a topological k-nearest neighbor graph with tunable exploration and exploitation dynamics with an adaptive repulsion parameter. This strategy is implemented within a simulated swarm of 50 agents with varying proportions of fast agents tracking a target represented by a fictitious binary function. Through these simulations, we are able to demonstrate an increase in the swarm's collective response level and target tracking performance by substituting in a proportion of fast buoys.