Team Arrival in Autonomous Racing - Roborace

Step inside the facility of Arrival's team | Roborace

author avatar

This article was first published on

Roborace is the competition for human + machine teams, using both autonomous-driven and manually-controlled AI cars. Race formats will feature new forms of immersive entertainment to engage the next generation of racing fans. Through sport, innovations in machine-driven technologies will be accelerated, tricking their way into society.  

Arrival's team use software and algorithms to test autonomous driving systems on race cars, competing against other teams from around the world. Development of the Robocar started in early 2016, with the first outing of a test vehicle, the so-called DevBot. DevBot 2.0 which is announced in late 2018, is all-electric and runs on the Nvidia DRIVE platform when in autonomous mode. 

It can be driven by a human or AI driver, meaning we can explore the relationship between man and machine for assisted and autonomous technologies. DevBot 2.0 has mainly four types of sensors, which are LIDAR, RADAR, Camera and GNSS. DevBot has 7 cameras uses them to collect 360° view around itself. 

This then allows the software to determine where it is. GNSS (Global Navigation Satellite Systems) that provides autonomous geo-special positioning with global coverage. This term includes GPS, GLONASS, Galileo and others. Most road cars fitted with the form GPS for their satellite navigation systems but GNSS on DevBot are military-grade, featuring ultra-fast update rate and enabling a location accuracy of 1cm. In future, Roborace cars will be fitted with ultrasonic sensors allowing them race wheel-to-wheel with greater precision.

Watch inside roborace.