Researching the potential for using decentralised, agent based, self-organised robots for testing and sorting second life batteries

该项目的目的是研究使用自动化电池拆解系统与人类合作，对电动汽车的电池进行重新用途，分类和分级，用于第二次生命应用。随着英国政府决定到2030年禁止新内燃机（ICE）汽车的销售，到2035年禁止混合动力汽车的销售，预计到2050年生产的电动汽车（EV）数量将大幅增加。因此，电池的数量将会很大，并且随着当前的自动化，这在存储，拆卸和回收方面提出了挑战。回收电池是耗时的，并且可能是昂贵和危险的。作为手动拆卸电池的替代方案，可以通过自动化过程来减轻触电、化学品暴露和电弧闪光的风险。然而，这是一个重大的挑战，因为电池组不是标准的，不同的车辆制造商选择不同的物理配置，电池类型和电池化学。目前，还没有已知的商业工厂进行自动电池拆卸以进行再利用应用。继ICE禁令之后，英国政府已投资2.46亿英镑用于法拉第电池挑战赛，以帮助新设施扩大规模，推进电池的生产、使用和回收。这项研究支持了这项工作，并研究了使用分散的、基于代理的、自组织的机器人来测试和分类第二次生命电池的可能性。这项工作将侧重于提高自动化系统的利用率，以实现更灵活和更高效的工作站。此外，还需要通过组合强化学习来探索使用人类和机器之间的共生关系，以利用人类的智能和机器快速、准确和可靠地处理数据的能力。

The aim of this project is to investigate the use of automated, battery dismantling systems with human collaboration to re-purpose, sort and grade batteries from electric vehicles for second life applications. With the decision by the UK government to ban the sale of new internal combustion engine (ICE) vehicles by 2030 and hybrid vehicles by 2035, it is expected that there will be a substantial increase in the number of electric vehicles (EV) being produced by 2050. As of this, the number of batteries coming through will be large and with the current automation this presents challenges in terms of storing, dismantling and recycling.To recycle a battery is time consuming and can be expensive and dangerous. As an alternative to manually dismantling the battery, it is possible to mitigate the risks of electrocution, chemical exposure and arc flash by automating the process. However, this is a major challenge because battery packs are not standard with different vehicle manufacturers opting for different physical configurations, cell types and cell chemistries.Currently, there are no known commercial plants undertaking automatic battery dismantling for re-use applications. Following the ICE ban, the UK government has invested £246 million into the Faraday battery challenge to help new facilities scale up and advance the production, use and recycling of batteries. This research supports that work and looks at the possibility of using decentralised, agent-based, self-organised robots for testing and sorting second life batteries. This work would focus on increasing the utilisation of automated systems to allow for more flexible and flow-efficient workstations. Moreover, there is a need to explore using symbiotic relations between humans and machines, through combined reinforcement learning, to utilise a human's intelligence with a machine's capability for processing data quickly, accurately and reliably.