At their algorithmic core, self-driving cars excel or are held back in their capacity for autonomy by their knowledge and understanding of the state of the environment surrounding the car. This, debatably, makes accurate sensor fusion and state estimation the most crucial element of a self-driving stack.
The state estimation challenge has been approached by various companies using a wide variety of methods, from different sensor suites to the fusion of real-time sensor data with mapping and navigational databases. How that data is gathered, processed, and leveraged represents the grand challenge of self-driving and the motivation for Quanser to initiate the Self-Driving Student Competition at the American Control Conference.
This initiative provides an excellent opportunity for students from around the globe to acquire leading-edge knowledge and develop critical problem-solving skills while also attracting and nurturing next-gen researchers.
The entire competition is designed as a combination of virtual and on-site competitions, giving student teams the opportunity to fully engage with QCar and its digital twin, QLabs Virtual QCar.
While the details of the competition may vary from year to year, it has always been Quanser’s endeavour to perpetually support students’ continued progress and innovation in the field of Self-driving!