Scholarlink

The growing integration of technology in health care has contributed to enhanced safety and quality of life of persons with mobility impairments. The following paper develops an IoT-based wheelchair fall detection and alert system that will improve the safety of wheelchair users. The system synchronizes the Arduino Uno microcontroller and the MPU6050 accelerometer-gyroscope module together with innovative AI algorithms for realistic fall detection. The proposed system integrates current information from the sensor, passes it through a particular model, and raises the alarm in case of conceivable falls to enable intervention. A comparison of the performance of two algorithms, Gradient Boosting and AdaBoost, fills in the analysis of F1 score and other characteristics. IoT connectivity integrated to the systems also ensures that some of the features of the system are monitoring and alerting from remote locations, making it a complete safety system. The work enhances knowledge in fall detection technologies and demonstrates how IoT supports the independence and well-being of wheelchair-bound individuals. Future developments include the enhancement of its accuracy as well as the possible addition of other functions for a user-friendly interface.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Smart IoT-Integrated Wheelchair: Fall Detection and Real-Time Alert System Using Machine Learning

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Aakanksha Mamgain,
  • Akshada Kendurkar,
  • Aswathy G. Krishnan,
  • Monica Bhutani,
  • Jyoti Gupta

摘要

The growing integration of technology in health care has contributed to enhanced safety and quality of life of persons with mobility impairments. The following paper develops an IoT-based wheelchair fall detection and alert system that will improve the safety of wheelchair users. The system synchronizes the Arduino Uno microcontroller and the MPU6050 accelerometer-gyroscope module together with innovative AI algorithms for realistic fall detection. The proposed system integrates current information from the sensor, passes it through a particular model, and raises the alarm in case of conceivable falls to enable intervention. A comparison of the performance of two algorithms, Gradient Boosting and AdaBoost, fills in the analysis of F1 score and other characteristics. IoT connectivity integrated to the systems also ensures that some of the features of the system are monitoring and alerting from remote locations, making it a complete safety system. The work enhances knowledge in fall detection technologies and demonstrates how IoT supports the independence and well-being of wheelchair-bound individuals. Future developments include the enhancement of its accuracy as well as the possible addition of other functions for a user-friendly interface.