Scholarlink

The CanSat uses an advanced ultra-light microcontroller, pressure and temperature sensors, a 3-axis accelerometer, a 3-axis gyro, a camera, GPS, an IR distance measuring sensor, and an RF communication module to communicate with a ground station PC. The design and navigation control of an advanced-level comeback CanSat, which is going to be launched to high altitude (between 2400 and 3658 m) using an amateur rocket from ground level, requires efficient control systems. In this paper, we present an advanced CanSat simulation and design methodology incorporating machine learning (ML) to enhance navigation, maneuverability, and resilience against sensor failures The simulation incorporates realistic sensor behaviors and malfunctions, such as gyroscopic drift, pressure sensor freezing, and temperature interference, enabling robust testing of ML-based solutions.

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

Design and Prognosis of CanSat Maneuver Systems Using Machine Learning

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Kouroush Jenab,
  • Saeid Moslehpour,
  • Edward Jhohan Marin Garcia,
  • Tyler B. Ward,
  • Lizeth D. C. Molina Acosta,
  • José Neftalí Torres Marin,
  • Jorge A. Ortega-Moody,
  • Jose Garcia

摘要

The CanSat uses an advanced ultra-light microcontroller, pressure and temperature sensors, a 3-axis accelerometer, a 3-axis gyro, a camera, GPS, an IR distance measuring sensor, and an RF communication module to communicate with a ground station PC. The design and navigation control of an advanced-level comeback CanSat, which is going to be launched to high altitude (between 2400 and 3658 m) using an amateur rocket from ground level, requires efficient control systems. In this paper, we present an advanced CanSat simulation and design methodology incorporating machine learning (ML) to enhance navigation, maneuverability, and resilience against sensor failures The simulation incorporates realistic sensor behaviors and malfunctions, such as gyroscopic drift, pressure sensor freezing, and temperature interference, enabling robust testing of ML-based solutions.