Currently, reducing effort or metabolic cost with the help of wearable robots is at the forefront of research by the wearable robotics and biomechanics community. Already a number of groups have achieved the reduction of metabolic cost using passive and active means [1, 2]. These mechanisms were tuned during extensive laboratory-based parameter adaptation, prior to achieving measurable metabolic cost reduction. An example is the passive ankle orthosis developed by Collins et al.,  which has achieved a reduction of 7.2% on the metabolic cost. The method employed discrete adaptation the flexibility of a clutched spring which undertakes part of the calf-muscle’s load during the push-off stage. In general, minimal reduction on the energy cost has been achieved which in effect challenges the potential price-to-benefit ratio. Furthermore, a significant variation in the outcome is observed when the mechanisms are used by different persons . Follow-up studies managed to further improve on the metabolic cost by applying real-time automatic customization, termed as Human-In the-Loop (HIL) optimization which alters the assistive force function given real time physiological measurements, the objective function being in most cases the metabolic cost. The groups that have applied this method have observed significantly higher improvement in performance. Also, as findings suggest, different criteria can be employed using the automatic customization method, such as learning and muscle activation levels [3, 4]. The new findings suggest that real-time customization can potentially lead to the next generation of exoskeletons that are employed for assistance, augmentation and rehabilitation purposes. This work aims to survey the current state of the art in the field of automatic customization of exoskeletons for assistance, augmentation and rehabilitation. It will identify and review the existing work, and specifically describe and compare the various tools, methods, and hardware that have been employed by the key research initiative. In addition to summarizing current scientific research questions, the work aims to establish a highly competitive 3-year research plan, which includes human and hardware resources, including the identification of potential key partners as well as multi-national and H2020 schemes for potential funding application. The primary focus of the work and plan will be lower-limb assistance using automatic customization methods. Application in simultaneous assistance and rehabilitation approaches will be considered. This work has been supported by COST Action CA16116 on Wearable robots for augmentation, assistance or substitution of human motor functions and is carried out in collaboration with Prof. Andreas Mueller from the Institute of Robotics, Johannes Kepler University, Linz.
 Collins, S. H., Wiggin, M. B., & Sawicki, G. S. (2015). Reducing the energy cost of human walking using an unpowered exoskeleton. Nature, 522(7555), 212–215. doi:10.1038/nature14288
 Mooney LM, Rouse EJ, Herr HM (2014) Autonomous exoskeleton reduces metabolic cost of human walking during load carriage. Journal of NeuroEngineering and Rehabilitation 11(1):80
 J. Zhang, P. Fiers, K. A. Witte, R. W. Jackson, K. L. Poggensee, C. G. Atkeson, S. H. Collins, Human-in-the-loop optimization of exoskeleton assistance during walking. Science 356, 1280–1284 (2017)
 Ding, Ye & Kim, Myunghee & Kuindersma, Scott & Walsh, Conor. (2018). Human-in-the-loop optimization of hip assistance with a soft exosuit during walking. Science Robotics. 3. eaar5438. 10.1126/scirobotics.aar5438.