Reputation management in open robot swarms through crypto-economy

In an open robot swarm task, cooperation is hindered by uncertainties in entities' willingness and local information restrictions. A proposed blockchain economy includes an information market to incentivize data sharing and penalize Byzantine attacks—spreading deceitful information with malice or incompetence. However, the protection mechanism faces vulnerability when robots receive colluded, locally shared information with heavy position noise. To address this, a global reputation management system is analyzed, using a metric proportional to robots' cooperative wealth visible on the blockchain, mitigating harmful effects of colluded, noisy information. Simulating a foraging mission where robots cooperate to locate and transport resources, a strategy is introduced to compare information source reputations, prioritizing higher reputations while rejecting noisy ones. Specific market rules harm Byzantines, extracting and redistributing wealth to cooperative robots, depleting deceivers of resources for market access. Robots can request debit wealth, exploiting Byzantines' borrowing tendencies and highlighting wealth disparities. Market stability and robustness are tested against two waves of Byzantine attacks, showcasing the swarm's defense against noisy and deceitful information. This pioneering study relies entirely on economic metrics, safeguarding against various forms of non-cooperation. Envisioning market theories and economic principles in swarm robotics leveraging blockchain technology's consolidation could advance the field.