Mobile Robotic Chemists For Advanced Formulations

阶段1使用现有设备在实验室中建立简化的工作流程，其中包括三个或四个独立的任务，这些任务将连续操作，例如剂量，混合和测量。这第一个工作流程和随后的任何其他工作流程将主要基于聚合物化学。最初的工作将围绕筛选实验室中的一系列潜在工作流程，以确定其适用性。正在研究的聚合物化学将具有工业应用，其中温度响应性聚合物或那些将增强机械性能的聚合物是将首先进行评估的有前途的工作流程。第二阶段下一步将使用移动的机器人，对于某些任务，使用静态机器人，分别完成工作流程中涉及的每项任务，开发代码以便于使用和轻松更改关键变量，如质量添加或混合时间。然后将开发代码，以允许在单个工作流中执行工作流中的多个任务，并再次能够像以前一样更改相同的关键变量。阶段3一旦机器人启动并运行，将在代码中实现故障和错误控制元素，以允许真实的时间解决机器人遇到的意外问题。沿着，还有可能将机器学习的某些方面纳入这种错误控制，以允许机器人开发自己的错误解决协议。第4阶段如果时间允许，将探讨如何在需要调整工作流程以反映方法上的微小变化时，以便于重新构建的方式构建代码。

Stage 1Set up a simplified workflow in the laboratory using existing equipment that involves three or four separate tasks that would operate in series, such as dose, mix and measure. This first workflow and any others that follow will largely be based around polymer chemistry. The initial work will revolve around screening a series of potential workflows in the laboratory to decide upon suitability. The polymer chemistry being looked at will have an industrial application with temperature-responsive polymers or those that will enhance mechanical properties being promising workflows that will be evaluated first. Stage 2The next step will be using the mobile robots, and for certain tasks the static robots, to do each of the tasks involved in the workflows separately, with the code being developed to allow for both ease of use and easy changing of key variables such as mass added or mixing time. The code will then be developed to allow for multiple tasks in the workflow to be carried out in a single workflow, once again with the ability to change the same key variables as before. Stage 3Once the robot is up and running there will be an element of fault and error control implemented into the code, to allow for real time solving of unexpected problems encountered by the robots. Along side this there is also the potential to incorporate some aspects of machine learning to this error control to allow the robots to develop their own error solving protocols. Stage 4If time permits, it will be explored as to how the code could be structured in a way that would allow easy reconstruction should the workflow need to be adapted to reflect small changes in the methodology.