Maintaining the situational awareness of an autonomous system responsible for delivering medication to patients in care homes

保持负责向疗养院患者提供药物的自主系统的态势感知

• 批准号: 2739286
• 金额: --
• 依托单位: University of York
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2022
• 资助国家: 英国
• 起止时间: 2022 至 无数据
• 项目状态: 未结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-2739286
• 关键词: Maintaining; situational; awareness; autonomous; system

Maintaining the situational awareness of an autonomous system responsible for delivering medication to patients in care homesProject My overall goal is to model the requirements of an autonomous system responsible for delivering medication to care home patients as a system of equations, which can then be used to predict times of future error in the system, so they can be prevented. The system should also function with a certain level of confidence, which can be calculated and hopefully minimized by adaptation of the system.-Create a system of equations representing the conditions needed to ensure the smooth operation of the autonomous system. I would start with a small number of conditions and then attempt to add more over the course of the project.- Create a method of prioritizing the equations. One way of doing this could be looking at which conditions would cause the most serious error if not met. Ideally, this would be done with input from a healthcare professional.- Each equation will be made up of several parameters. Monitor these and generate a safe range for each parameter, where there is no risk of critical error.- From these safe ranges of each parameter, predict the time at which the next fatal error could occur. This way, if known, it could likely be prevented.- Calculate a bounded probability of error for the system. This could potentially be done by hypothesis testing which is updated each time the model produces a new future risk.

我的总体目标是将负责向养老院患者提供药物的自主系统的需求建模为一个方程系统，然后可以使用它来预测系统中未来错误的时间，从而可以预防错误。系统还应该以一定程度的置信度运行，这个置信度可以计算出来，并希望通过系统的自适应将其最小化。-创建一个方程组，表示确保自治系统顺利运行所需的条件。我会从少量条件开始，然后在项目过程中尝试添加更多条件。-创建一个优先排序方程的方法。这样做的一种方法是查看如果不满足哪些条件会导致最严重的错误。理想情况下，这将在医疗保健专业人员的输入下完成。-每个方程将由几个参数组成。监视这些参数并为每个参数生成一个安全范围，在此范围内没有发生严重错误的风险。—根据每个参数的安全范围，预测下一个致命错误可能发生的时间。这样，如果知道，就有可能预防。—计算系统的有界误差概率。这可以通过假设检验来实现，每次模型产生新的未来风险时都会更新假设检验。