Abstract
The growing popularity of electric scooters has resulted in surfacing individual and public safety concerns associated with riding them. Many of these safety issues are associated with unsafe riding behaviors. While the terms of the riding agreements of scooter companies explicitly prohibit potentially unsafe riding behaviors, automatic and dynamic detection of unsafe riding is not easily possible simply from the location and speed data typically collected by the scooters. This disclosure proposes fitting an electric scooter with a weight-detecting mechanism that allows dynamic determination of riding behavior that indicates unsafe and/or impermissible use. Pattern recognition and/or machine learning is applied to process aggregated weight measurements to identify riding patterns in different riding positions and situations. The scooter provider can then take appropriate action to deter unsafe or non-compliant riding, thus incentivizing safe and responsible riding habits and improving street safety.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Terrell, Gabriel C., "Characterizing electric scooter riding behavior to detect unsafe and non-compliant use", Technical Disclosure Commons, (June 13, 2019)
https://www.tdcommons.org/dpubs_series/2280