Barcelona to trial AI-powered traffic lights for better traffic management

Kapsch estimates that its system could reduce overall congestion by up to 10% in traffic situations where speeds are below 10 km/h. This is a forecast that still needs to be tested on the streets of Barcelona.

The city is preparing a four-month pilot with two artificial intelligence solutions that will operate at several intersections. Real-world trials are scheduled for late 2027.

Two ways to adjust traffic lights

The Matcab proposal, from the Austrian company Kapsch, will monitor traffic and adjust the timing of each traffic light according to the needs of the intersection. The system will learn through simulations and can activate different objectives, prioritising pedestrians, public transport, or emergency vehicles.

According to the company’s forecasts, active mobility could see waiting times reduced by up to 5%. Public transport journey times would fall by 3%. Kapsch also estimates a reduction of up to 10% in overall congestion in slow traffic, below 10 km/h. The system can be integrated into existing infrastructure and prevent queues from shifting from one intersection to another.

Kapsch’s project has a budget of 358,185 euros. The Ayuntamiento will contribute 100,000.

A digital twin to coordinate intersections

The second solution is called MOB-IA and proposes creating a digital copy of the road network. It will receive real-time data from sensors and V2X connected vehicle systems to anticipate how traffic evolves and coordinate the traffic lights along a corridor. The proposal is attributed to Acsia; among the companies selected to implement the solutions is Acisa.

Multi-agent artificial intelligence will allow the cycles to be adjusted with a chosen priority, such as favouring pedestrians, public transport, or reducing emissions. Its forecasts suggest cutting regular traffic time by between 10% and 20%. It also estimates between 15% and 25% fewer stops, as well as a 10% to 20% reduction in travel time variability.

MOB-IA requires minimal intervention in the infrastructure and will allow the current system to be maintained as a reference. Its budget amounts to 177,865 euros, of which the council will finance 100,000.

A timeline linked to changes on the Diagonal

The roadmap for incorporating AI into traffic light regulation has been designed by the Movilidad del Ayuntamiento and the Universitat Politècnica de Catalunya (UPC). The projects will start at the end of 2026, and the experimental phase will serve to evaluate the technologies before considering a possible larger-scale rollout.

The trials will coincide with the works to connect the tram along the Diagonal, which will require reorganising traffic on one of the city’s busiest axes. The selection of Kapsch y Acisa was announced during Tomorrow Mobility, part of the Smart City Expo World Congress, from among 26 candidates, 19 of them international.

TMB and ATM join the laboratory

The event also formalised the incorporation of Transports Metropolitans de Barcelona (TMB) and the Autoritat del Transport Metropolità (ATM) into the Barcelona Innova Lab Mobility (BILM), an initiative of the Ayuntamiento y Fira de Barcelona with participation from the Fundación Mobile World Capital. The ATM is a public consortium involving the authorities responsible for public transport in the Barcelona area.

TMB, operator of the metro and urban bus, joins after collaborating in previous challenges focused on improving the bus service. Successful pilot tests include AI cameras to detect improper use of bike lanes and technology at bus shelters to collect information and improve the service.

Traffic regulation is added to other municipal mobility issues, such as disputes over fines. Manuel Valdés, general director of the ATM, argued that innovation must leave the laboratory and be useful for users.