Automated Feedback Selection for Robot-Assisted Training

Abstract

Robot-assisted training can be enhanced by using augmented feedback to support trainees during learning. Efficacy of augmented feedback is assumed to be dependent on the trainee's skill level and task characteristics. Thus, selecting the most efficient augmented feedback for individual subjects over the course of training is challenging.We present a general concept to automate feedback selection based on predicted performance improvement. As proof of concept, we applied our concept to trunkarm rowing. Using existing data, the assumption that improvement is skill level dependent was verified and a predictive linear mixed model was obtained. We used this model to automatically select feedback for new trainees. The observed improvements were used to adapt the prediction model to the individual subject. The prediction model did not over-fit and generalized to new subjects with this adaptation.Mainly, feedback was selected that showed the highest baseline to retention learning in previous studies. By this replication of our former best results we demonstrate that a simple decision rule based on improvement prediction has the potential to reasonably select feedback, or to provide a comprehensible suggestion to a human supervisor. To our knowledge, this is the first time an automated feedback selection has been realized in motor learning.

When a robot teaches humans: Automated feedback selection accelerates motor learning

“…With respect to the velocity error and the spatial error entire cycle, both groups showed improvements and reached absolute performance levels similar to those in previous studies of the same task (13,16,31). For this reason, the subdivision of visual feedback into the four rowing phases and the novel multimodal feedback combinations proved to be suitable for learning the rowing movement.…”

“…In addition, normalization by the smallest errors observed in previous studies allowed us to reasonably scale the magnitude of different error metrics to each other based on effectively reachable performance. Therefore, no manual tuning of error metrics had to be performed, and only minimal prior knowledge or data were required compared with other approaches relying on large datasets (13). Although the velocity error was dominant for 50% of all trainings and the spatial error at release was not dominant for a single training, the sequence of dominant errors was rather individual (Table 1).…”

“…In addition, normalization by the smallest errors observed in previous studies allowed us to reasonably scale the magnitude of different error metrics to each other based on effectively reachable performance. Therefore, no manual tuning of error metrics had to be performed, and only minimal prior knowledge or data were required compared with other approaches relying on large datasets (13).…”

“…We adopted the approach of incorporating human expertise in a robotic trainer by addressing the largest error out of different performance metrics by a dedicated type of augmented feedback. The movement task was to learn an oar blade trajectory in trunk-arm sweep rowing, as in our previous study (13). The oar had to be moved according to a predefined velocity profile, i.e., fast in the water and slow in the air phase.…”

“…Observed error reductions were used to adjust the model to the individual participant. Our predictive concept demonstrated a reasonable selection of training conditions featuring different types of feedback (13). However, a considerable amount of data was required to develop and justify the underlying model.…”