In the military, police, security companies, and shooting sports, precision shooting training is one of the most important courses. To improve precision shooting performance, trainees will consume a large number of cartridges and an immense amount of professional coaches` time -both could cost a lot. We have designed an augmentedbiofeedback system based on machine learning techniques to reduce costs and shorten training time. Our system works in two different scenarios: concurrent feedback and terminal feedback. An accuracy prediction model for precision shooting based on random forest (RF) works in concurrent feedback that provides the user with real-time audio feedback to suspend the current shot, if the model predicts a poor result. The terminal feedback provides information about the statistical values of the shot accuracy, possible errors that may occur during the current training, and gives suggestions for improvement during the next shooting episode. Experimental results show that not only does the proposed RF model achieve the accuracy of 91.27%, higher than any of the existing reference models, but also our system has the potential to reduce cartridge consumption and the time of trainers
© Copyright 2020 IPSI BgD Transactions on Advanced Research. IPSI BgD. Alle Rechte vorbehalten.