Autonomous Driving Challenge

The Autonomous Driving Challenge is a university program in which participants develop fully automated driving functions in a state-of-the art scaled vehicle.

The teams meet at a final event after eight months of developing the software solution.

Challenges include track localisation and navigation, perception of traffic signs, pedestrian crossings and traffic lights, and finally car manoeuvres such as parking and overtaking static and dynamic obstacles.

Software development is precisely the biggest challenge facing autonomous cars. Thus, despite using a small scale model, the technology and the software structures are the same as a full-scale car. With the Autonomous Driving Challenge, we offer students the possibility to write part of this future mobility concept!

The Competition

The Autonomous Driving Challenge is an annual competition aimed at students with a technical background in robotics who are interested in developing the automotive technology of the future. Participants program fully automated driving functions and the necessary software architectures.


They work with 1:8 scale vehicles, developed exclusively for the competition. The cars are equipped with a LIDAR sensor, ultrasound, cameras, a high-performance NVIDIA graphics card, a CPU and ROS libraries.

Thanks to in-house developed software, the models should be able to drive themselves according to all rules and avoid obstacles.
The project starts at the beginning of the year, when the selected teams receive the model vehicle. The organisation provides teaching materials, training sessions and technical talks by robotics experts to ensure that all participants have a common starting point.

The teams will have eight months to develop their software solution, with the support of a university tutor, which will eventually allow the cars to navigate autonomously around a track. During this time, they will be mentored by the organisation and will perform intermediate tasks.

The final test will give the students the opportunity to demonstrate their knowledge and efforts. The vehicles will have to overcome a series of challenges on a track design and built specifically for the event. They will test their cars in different scenarios, related to the requirements of a real environment.

Previous Editions

Third Edition

In November 2021 four teams from Spanish universities met at the Tomorrow Mobility Congress of the Smart City Expo in Barcelona. The CUPRA Formentor model cars navigated the streets of a Barcelona-based circuit for 3 days. As side events we had numerous speakers from leading companies in AV and 5G, we showed the operation of a car driving software aimed at people with reduced mobility and finally we had the competition live on the InfoK channel.

Previous Editions

Second Edition

The second edition of the Autonomous Driving Challenge was held in 2018 at the event #Imperdible_03 in Madrid. This expo is organized and sponsored every year by COTEC, a private foundation whose mission is to promote innovation in Spain. The competition lasted two days and included many lectures and workshops about robotics and programming.

Previous Editions

First Edition

The Autonomous Driving Challenge was first held in 2017, as the first public challenge on autonomous driving in Spain. Over 50 students from 7 spanish universities around the country participated during the Smart City Expo in Barcelona, which was attended by over 18 thousand people.

Participants of the CARNET Autonomous Driving Challenge will receive one model vehicle per team. The hardware platform is fully sensorized with high tech components.

Power Supply

The car’s power is supplied by two separate energy circuits with a 22.2 V 6-cell battery with a capacity of 5,200 mAh, which allows the car to operate in high performance mode for over an hour. For the measuring technology and the computer, an external power supply unit with a voltage range of 6-24 V can be used.


The Formentor miniatures are equipped with a miniTX board. This includes an Intel Core i3 processor, 8 GB RAM, a fast 128 GB M.2 SSD hard drive and an NIVIDIA GeForce GTX1050Ti graphics card. In addition to two Gigabit Ethernet ports, the board also has several USB 3.0 interfaces and a USB-C port. Furthermore, a Bluetooth and a WLAN module (IEEE 802.11ac) is available. The communication interface with the vehicle has an additional Arduino micro, so that every single cell voltage of the batteries can be monitored.


The propulsion engine of the prototype is a high-torque brushless motor with cruise control that allows controlled forward and reverse driving, as well as braking in both directions, even at low speeds. The steering system is provided by a digital servomotor used in model racing. It has a high actuating power at low operating times, which ensures precise, fast steering action.


The sensor set is pretty close to a real car sensor set. The whole pack consists of the following components:
RPLIDAR A2 (front, > 180 field of view, detection range < 6 m, update rate 10 Hz).

  • 130° mono video camera (front).
  • 80° mono video camera (back).
  • Ultrasonic sensors (three in the back, one on the left, one on the right, detection range < 4 m, update rate 40 Hz).
  • 9-axis motion tracking sensor to measure acceleration and angular rates.
  • Wheel speed sensors to measure the speed of the car
  • Organized and delivered by: