Congestion

CAVs offer great potential for substantial improvements in network performance, study finds

A study by the UK Department for Transport finds that CAVs offer, particularly in high-speed, high-flow situations, great potential for substantial improvements in network performance. However, such substantial benefits may not be achieved until there exist high levels of connectivity and automation, due to limitations by other vehicles' behaviour. This leads to the suggestion of a tipping point – the proportion of enhanced vehicles required before benefits are seen. This work suggests this may be between 50% and 75% penetration of CAVs.

The study also suggests low speed urban areas may benefit most from low-tech driver assistance capability. Results from the urban model suggest initial benefits to delay of more than 12% with a 25% penetration of CAVs, rising to 30% with a fully automated vehicle fleet.

To access the full paper, click here.

How autonomous vehicles could relieve or worsen traffic congestion

With the number of cars on the road globally forecast to double from 1.1 billion to 2 billion in the next 15 years, traffic congestion continues to be a multi-trillion-dollar drain on the global economy. This whitepaper demonstrates how the introduction of autonomous cars will not in itself fix the problem of congestion. In fact, unless the industry takes action now, there are many reasons why autonomy would actually worsen traffic conditions in the immediate future.

What kind of action is required? As we argue, what is needed is a concerted effort to shift away from ‘each-to-their-own’ autonomy (where each car is only responsible for itself), and towards Collaborative Autonomous Cars. Only by collaboratively collecting data can the industry and policy makers truly understand the holistic impact of autonomy. Only by sharing rich vehicle sensor data between cars can we build bridges between autonomous and non-autonomous cars. Only by enabling collaborative management of autonomous fleets can societies truly optimize road networks. And, only by breaking down data silos and connecting vehicle data, road network data and infrastructure data such as stop light information can we truly realize the full benefits of autonomy.  

Finally, only when the industry finds a way to collaborate in parallel to designing cutting-edge technologies, can autonomous cars be considered to be on the path towards sustainable success.

To read the full paper, click here.