Development of an Inertial Measurement Unit with Fast Response Time
|Supervisor:||Horst Hellbrück , Grigori Goronzy|
Inertial Measurement Units (IMUs) have a variety of uses, including robot and AUV pose determination. They form a basis for self-balancing of differential steering mobile robots which are used in the LOCIC project and can also be used to assist odometry. An IMU that focuses on fast response time is required for that.
A combined pose estimation suitable for balancing and odometry assistance needs to be developed and evaluated. IMUs combine outputs of inertial sensors. Each type of sensor has various weaknesses and strengths, and it is typically required to combine raw readings of sensors. Combining values with a Kalman filter, complementary filter or DCM results in output with fast response time, improved accuracy and low drift.
- Getting familiar with accelerometer, gyroscope and magnetometer sensors and the basics of sensor fusion
- Implementation of a driver for the MPU-9250 chipset to calibrate sensors and to read and correct sensor values
- Implementation and tuning of 6-DOF (accelerometer/gyroscope) or 9-DOF (accelerometer/gyroscope/magnetometer) sensor fusion with a proven algorithm
- Evaluation of fused data response time, drift and accuracy
- Good knowledge of C programming
- Basic knowledge of microcontrollers and digital electronics
- Solid knowledge of digital signal processing is a plus