Formula Student Driverless competition
The Formula Student Driverless competition includes a number of different challenges: braking, acceleration, skidpad testing and a track drive.
Chalmers University of Technology, a Swedish university which focuses on research and education in technology, participated to the 2018 edition. The Chalmers team has equipped their driverless car with the Ellipse2-N, a miniature inertial navigation system with embedded GNSS receiver.
Robust and Stable Inertial Navigation Sensor
We were very happy with the performance and robustness of the sensor. The sensor did not have any apparent errors during the hundreds of hours of testing we performed.
We did not experience any drifting for the inertial sensors and we were especially impressed by the excellent yaw rate estimation. The GNSS system was also very robust and stable and we never experienced any problems with lacking GNSS coverage and we always had a good number of satellites.
Easy-to-integrate thanks to the C library
The Ellipse2-N was very easy to integrate into our own software framework by using the provided C library. Our framework used microservices hosted in Docker containers where the data from the sensor was read by one microservice.
This meant that it was important that the microservice could be built automatically and seamlessly. By the provided library it was simple to include the necessary code into the Docker image and build it together with our own code interfacing the library.
If there would only have been a provided binary instead we would have had to read and parse the raw sensor data ourselves so the library was very useful for our case. The examples and documentation provided made it very easy to use the library and interface the Ellipse2-N into the microservice we needed.
Single or Dual Antenna for Such Conditions?
If a single antenna Inertial Navigation System (INS) like the Ellipse2-N provides very accurate orientation and navigation data, a dual antenna INS such as the Ellipse2-D allows a faster initialization, even in a static position.
Such features are to be considered when choosing your INS. Find more information about « how to choose the best heading method ».
We did not experience any drifting for the inertial sensors and we were especially impressed by the excellent yaw rate estimation
Ellipse-N
Single Antenna RTK GNSS
- 0.05° Roll and Pitch (RTK)
- 0.2° Heading (RTK high dynamics)
- 5 cm Real-time Heave
- 1 cm RTK GNSS Position
- Raw Data for Post-Processing
- Extremely small OEM module
- Full Development Kit
- ROS Driver