A minimal biologically-inspired algorithm for robots foraging energy in uncertain environments

Abstract

This work details the design and simulation results of a bioinspired minimalist algorithm based on C. elegans, using autonomous agents to forage for attractant energy sources. The robotic agents are energy-constrained and depend on the energy they forage to recharge their batteries, which is significant as the foraging task is one of the canonical testbeds for cooperative robotics.The algorithm consists of 6 input parameters which were simulated and optimised in 9 unbounded environments of varying difficulty levels, containing attractant sources which robots would then have to forage from to maintain energy levels and survive the entirety of the simulation.The robots running the algorithm were then optimised using Evolutionary Algorithms and the best solutions in all 9 environments were categorized with the use of clustering techniques. The clustering results highlighted the different strategies which emerged. Ultimately across the 9 environments, 6 different strategies have been identified. The results demonstrate the applicability of the proposed algorithm to localise attractant sources and harvest energy in different scenarios using the same core algorithm.

RepAtt: Achieving Swarm Coordination through Chemotaxis

“…Furthermore, the robots sense and broadcast these signals selectively depending on the state they are in, thereby creating a dynamic sensory landscape. These properties make our approach different from the biological foundations of our algorithm and its previous robotic implementations in [17] for localising sound source and in [19], where swarm robots foraged energy from light spots in their environment.…”

Swarm Foraging Under Communication and Vision Uncertainties

“…RepAtt is inspired by the chemotactic search behaviour observed in micro-organisms such as the Escherichia coli bacterium and Caenorhabditis elegans nematode. The novelty of RepAtt lies in the use of the foraging robots as sources of signals whose intensity degrades with distance, unlike other implementations that used immobile signal sources [32], [33]. Neighbouring robots then sense the change in intensity of these signals and use them to perform chemotactic search for good areas in which to forage.…”

Swarm Foraging Review: Closing the Gap Between Proof and Practice

“…Andrade and Boyle [33] investigated an energy constrained model of foraging in which robots use the energy from foraged resources to continue foraging. Evolutionary algorithms were used to tune a 6-parameter model over several environments.…”