Biological inspired control and machine learning for clinical rehabilitation and engineering systems

解释人类如何控制运动和姿势长期以来一直是科学家和工程师的兴趣，关于人类运动控制的确切性质仍然存在争议，以及这与他们的适应和学习能力如何相关。近年来开发的一种方法是基于开环预测控制和间歇闭环控制的组合，称为间歇预测控制。该项目与正在进行的控制理论基础研究有关，并应用于大数据科学（EPSRC项目EP/R 018634/1 "用于复杂、计算和数据密集型分析的闭环数据科学“）。从理论发展中获得的对人类控制的见解将用于康复方法的应用。由于康复是一个再学习的过程，了解人类如何学习提供了一个坚实的基础，为设计新的rehabilitation approaches.Aims和objectives：本项目的目的是调查应用程序的适应和学习的基础上间歇性控制的概念，在神经康复。具体目标将是：调查机制，为人类一样的适应和学习，评估机器学习可以在自适应间歇控制结构中发挥的作用，探索间歇控制的潜力，以改善联合人机交互，重点在神经康复中的应用。新奇的研究方法：间歇控制是一个相对较新的控制概念，最初是基于模型预测控制。它在适应和学习方面的特性目前还没有得到很好的理解。在这个项目中采取的新方法将是发展的理论特征的理解为工具，可用于在临床康复方面的再学习。符合研究理事会的战略和研究领域：重点领域是医疗技术，特别是优化治疗。任何参与的公司或合作者：临床评估将与格拉斯哥伊丽莎白女王大学医院的伊丽莎白女王国家脊柱损伤单位合作进行。

Brief description of the context of the research including potential impact:Explaining how humans control movement and posture has been of interest to scientists and engineers for a long time, with controversy remaining about the exact nature of human motor control, and how this is related to their ability to adapt and to learn. One approach which has been developed in recent years, is based on a combination of open-loop predictive control and intermittent closed loop control, termed intermittent predictive control. This project is related to ongoing basic research in control theory, with applications in Big Data science (EPSRC project EP/R018634/1 ``Closed-Loop Data Science for Complex, Computationally- and Data-Intensive Analytics''). Insights into human control gained from theoretical developments will be used in applications in rehabilitation approaches. Since rehabilitation is a process of relearning, understanding how humans learn provides a solid basis for designing novel rehabilitation approaches.Aims and objectives:The aim of this project is to investigate the application of adaptation and learning based on intermittent control concepts in the context of neuro-rehabilitation. Specific objectives will be:Investigate mechanisms for human-like adaptation and learning, Evaluate the role machine learning can play in an adaptive intermittent control structure, Explore the potential of intermittent control to improve joint human-machine interaction, focusing on applications in neuro-rehabilitation.Novelty of the research methodology:Intermittent control is a relatively new control concept which is originally based on model predictive control. Its properties in the context of adaptation and learning are currently not well understood. The novel approach taken in this project will be to develop the understanding of theoretical characteristics into tools which can be used for relearning in a clinical rehabilitation context. Alignment to Research Council's strategies and research areas:Focus areas are Healthcare Technologies, specifically Optimising treatments.Any companies or collaborators involved:Clinical evaluations will be conducted in collaboration with the Queen Elizabeth National Spinal Injuries Unit at the Queen Elizabeth University Hospital, Glasgow.