Testing on public roads

Catalonia Living Lab provides for testing of connected and automated vehicles on Catalan public roads. Testing on Catalan public roads is not limited to predefined routes and areas: customized testing solutions on public roads can be designed in case of specific requests.

Selection criteria

The routes and areas presented here, are based on a comprehensive market research, including a series of interviews with leading car manufacturers and their suppliers. Based on that research, the following criteria are applied for the selection of routes and areas:

+ Presence of sensor blockers

Obstacles blocking the view of sensors, are challenging for any vehicle trying to find its way autonomously. Examples of sensor blockers: parked vehicles, buildings, bridges, trash deposits, fences, barriers, vegetation, trucks, busses, hills, direct sunlight or other road users.

+ Presence of sky blockers

Obstacles blocking the sky, can prevent sunlight or satellite signals to reach the autonomous vehicle. This can lead to varying lighting conditions and difficulties in vehicle localization (due to reduced GNSS signals). Example of sky blockers are: bridges, tunnels, sky scrapers, canyons, trees or any other overhead structures.

+ Presence of road obstacles

Unusual or unexpected obstacles on the road, can be difficult for an autonomous vehicle to deal with. Perception technologies might not be able to understand what the vehicle is facing, which can result in undesired behavior. Examples of obstacles on the road: construction works, accident scenes, police control, trash, trash deposits, animals, trees, stones, soil, groups of people, cones, fences, barriers, speed bumps, potholes, floodings, bus at bus stop, tram or railroad crossing.

+ Presence of communication issues

Any issues related to wireless communications between a vehicle and the rest of the world, can create problems for autonomous driving functions that rely on that data. This communication can have issues due to fading, noise, network switching, antenna switching or border crossing. Fading is the attenuation of a signal (damping), for example caused by reflections on buildings or large obstacles in between the sender and the receiver.

Noise is any type of disruption that interferes with the transmission of data from the sender to the receiver. Examples of objects or locations that can cause communications issues are: bridges, tunnels, buildings, vegetation, sky scrapers, overhead structures, other transmitting devices, canyons, borders and remote or sparsely populated areas.

+ High road user density

Dynamic scenarios including many other road users, are challenging for autonomous vehicles. Especially in case of vulnerable road users (VRUs), becasue mistakes can easily lead to injuries or fatalities. Many objects needs to be recognized and their trajectories estimated, based on which the desired actions should be taken. Areas in which high density road users can be expected are: city access roads, city centers, ring roads, congestion areas and touristic areas.

+ High traffic sign density

Autonomous vehicles might struggle to correctly detect or understand certain traffic signs. For example when signs are electronic, unusual, damaged, (partially) blocked or include text. The more traffic signs together, the more challenging it becomes for an autonomous vehicle to correctly interpret and obey each one of them. Examples of areas with high density traffic signs are: city access roads, city centers and around intersections.

+ Varying road characteristics

Automated driving functions might be tested for a limited number of road characteristics. Varying the characteristics of a road can therefore be challenging for autonomous vehicles. Examples of road characteristics are: widths, edges, gradients, curve radii, surface materials, line types and barriers.

+ Different intersections

Intersections can be challenging in many ways for an autonomous vehicle. Its trajectory will be cut by other road users with which it needs to interact safely. Traffic signs need to be detected and respected. Sensor views can get blocked by objects or other road users. Dimensions and allowable speeds can vary. Examples of different intersection types are: T, X, Y or roundabouts.

+ Parkings

One of the main automated driving functions of importance for the near future, relates to parking a vehicle. The biggest challenge is indoor parking, since it includes steep ramps, structures blocking sensors, reduced lighting and no GNSS signals. Examples of parking types are: indoor parking garages or outdoor parking lots.

+ High speed road entries and exits

Other automated driving functions of importance for the near future, relate to highway driving. Entering or exiting a highway (or any other high speed road), is one of the challenges that autonomous vehicles have to master, correctly incorporating themselves into traffic flow. Entries and exits come in different shapes and sizes.

Scoring

Based on the criteria explained above, routes and areas are selected across Catalonia. Each route or area is assigned a score between 1 and 10 for each criterion. The higher the score, the more challenging for a connected or automated vehicle to drive on that route or area. As can be seen in the following image, the combined set of selected routes and areas is highly challenging on all criteria.