Mealin S, Foster M, Cleghern Z, et al. Using Inertial Measurement Unit Data for Objective Evaluations of Potential Guide Dogs. ACI '20: Proceedings of the Seventh International Conference on Animal-Computer Interaction. 2021;2:1-11. https://doi.org/10.1145/3446002.3446064.
This study explores the use of inertial measurement unit (IMU) data, specifically from accelerometers and gyroscopes, to objectively assess guide dog candidates during standard temperament evaluations. A neural network was developed to predict evaluation scores typically assigned by experienced human observers based on the IMU data collected during these assessments. The model achieved an accuracy of approximately 93%, demonstrating the potential for automated, objective evaluation methods in the guide dog training process.
Comments