Neurorobotik
Current Topic
Neurorobotik 

AI-Driven Torque Estimation in Robotic Systems: Leveraging Sensor Fusion for Enhanced Accuracy

AI generated image showing some abstract view of a motor with different diagrams in the background as well as brain-like structures.AI generated (Gemini)

Modern robotic systems rely on accurate torque estimation for reliable operation. This thesis focuses on applying AI/ML techniques to estimate motor torque using sensor data recorded from the motor under various operating conditions, such as changes in load, velocity. The dataset includes time-series measurements like velocity, torque, encoder angles, motor current, and temperature, also torque values from an external reference sensor serving as the reference torque (Ground truth). The goal is to use ML AI techniques that can predict torque values closely matching those of the reference sensor.

Advisor:

Requirements:

  • Python, MATLAB, basic understanding of machine learning and AI concepts.
  • Familiarity with physical system modeling or motor control can be helpful but not mandatory.