From industrial exoskeletons to implantable medical devices, robots that interact closely with people are poised to improve every aspect of our lives. Yet designing these systems is very challenging; humans are incredibly complex and, in many cases, we respond to robotic devices in ways that cannot be modelled or predicted with sufficient accuracy. A new approach, human-in-the-loop optimization, can overcome these challenges by systematically and empirically identifying the device characteristics that result in the best objective performance for a specific user and application. This approach has enabled substantial improvements in human-robot performance in research settings and has the potential to speed development and enhance products. In this Perspective, we describe methods for applying human-in-the-loop optimization to new human-robot interaction problems, addressing each key decision in a variety of contexts. We also identify opportunities to develop new optimization techniques and answer underlying scientific questions. We anticipate that our readers will advance human-in-the-loop optimization and use it to design robotic devices that truly enhance the human experience.

@en

Exoskeletons often allow limited movement of the ankle joint. This could increase the chance of falling while walking, particularly on challenging surfaces, such as lateral inclines. In this study, the effect of a mobility limiting ankle brace on gait stability in the frontal plane was assessed, while participants walked on lateral inclines. The brace negatively affected gait stability when it was worn on the leg that was on the vertically lower side or ‘valley side’ of the lateral incline, which would indicate an increased risk of falling in that direction.

@en

Children with cerebral palsy (CP) often show excessive knee flexion during walking. To counteract knee flexion, Ankle-Foot Orthoses (AFOs) are frequently prescribed. The effectiveness of AFOs is dependent on their mechanical characteristics, e.g. stiffness [1] and alignment [2]. The Roll-over Shape (RoS) has been proposed as a parameter to quantify the AFO’s mechanical behaviour and alignment [3]. The RoS has however never been investigated in children with CP walking with or without AFOs.@en

Background Knowledge on lower extremity strength is imperative to informed decision making for children with cerebral palsy (CP) with mobility problems. However, a functional and clinically feasible test is not available. We aimed to determine whether the squat test is suitable for this purpose by investigating test performance and execution in children with cerebral palsy and typically developing (TD) peers. Methods Squat test performance, defined by the number of two-legged squats until fatigue (max 20), was assessed in twenty children with bilateral CP (6–19 years; gross motor function classification system I–III) and sixteen TD children (7–16 years). Muscle fatigue was assessed from changes in electromyography (EMG). Joint range-of-motion and net torque were calculated for each single squat, to investigate differences between groups and between the 2nd and last squat. Findings Fifteen children with CP performed < 20 squats (median = 13, IQR = 7–19), while all TD children performed the maximum of 20 squats. Median EMG frequency decreased and amplitude increased in mm. quadriceps of both groups. Ankle and knee range-of-motion were reduced in children with CP during a single squat by 10 to 15°. No differences between 2nd and last squat were observed, except for knee range-of-motion which increased in TD children and decreased in children with CP. Interpretation Squat test performance was reduced in children with CP, especially in those with more severe CP. Muscle fatigue was present in both children with CP and TD peers, confirming that endurance of the lower extremity was tested. Minor execution differences between groups suggest that standardized execution is important to avoid compensation strategies. It is concluded that the squat test is feasible to test lower extremity strength in children with CP in a clinically meaningful way. Further clinimetric evaluation is needed before clinical implementation.

@en