Autonomous cars will cause traffic jams

It will take much more for robotic vehicles to spread on the roads to change and improve the traffic situation.

Waymo has been testing its autonomous vehicles on Phoenix roads in recent years and it was common for them to brake or stop for seemingly no reason. Software and artificial intelligence experts have finally found the cause of the problem: the birds. The vehicles regarded the animals as obstacles and braked each time a bird approached them. Waymo finally solved the problem by further developing the algorithm.

This example perfectly shows that developers still have a lot to do, as it also highlights that autonomous cars follow traffic rules so rigidly that they behave completely differently than people. This is because neural networks are prepared as a precaution, that is, to maintain a greater tracking distance, to accelerate more slowly, or to stop at intersections for a longer period of time. Bottom line: these systems travel in extremely defensive ways.

Simulations by the PTV software company have shown that this practice will have consequences for everyday traffic. Traffic can already be disrupted if the proportion of autonomous cars is 20 percent. If it is 50 percent, the proportion of plugs can double. Jochen Lohmiller, PTV project manager for it He toldthat these things do not bode well for cities.

Researchers see the future in so-called digital twins. This means that roads are monitored with cameras, radars, LiDARs and other sensors, and then anonymized (without license plates) data is processed using artificial intelligence to create a digital imprint of traffic. This real-time 3D map allows you to plan routes more efficiently and increase traffic throughput. Alois Knoll, head of the Department of Robotics, Artificial Intelligence and Real-Time Systems at the Technical University of Munich, said it was all extremely important for the sustainability of the road network.


Real-time traffic monitoring and analysis would be required

Autonomous vehicles can improve traffic, but only if their proportion is well over 50 percent. This is because the cars connected to the network can coordinate their movements with each other, adjust to an optimal speed and calculate or narrow down the following distances. According to transport experts, nothing is better for handling traffic than vehicles moving smoothly. The problem lies in the interaction between autonomous cars and people-driven cars.

According to Knoll, people simply react differently to different situations. Traffic lights, for example, have an accordion effect. Drivers start not at once but intermittently, the distances between vehicles are getting bigger and fewer cars can pass than a green light. This does not help the transition, which could take 10 years even if all cars and trucks on the market today are SAE Level 5, meaning fully automated.

The situation is urgent, there is not much time for transport engineers because autonomous driving is no longer a vision. With the new Mercedes S-Class, you can drive at speeds of up to 60 km / h on a congested motorway without the driver having to hold the steering wheel. This is called Level 3 and is ready for series production. Here, under certain circumstances, the driver gives up control of the vehicle, but must be prepared to intervene again at any time. Implementing fully autonomous leadership at Level 4 is more technologically complex, but progress with people is even more confusing, as many test projects around the world have shown. At the highest level 5, the government and leader are completely unnecessary. It will probably be many years before this level of autonomy becomes part of everyday life.

Based on PTV simulations, if 20 percent of all vehicles in traffic are autonomous, the average speed will decrease from 23.5 to 20.9 kilometers / hour. In Cologne, for example, the PTV found an average speed of 58 km / h for human drivers. This is faster than allowed in populated areas. Autonomous vehicles, on the other hand, follow the standard at exactly 49 km / h. But if the rate is 50 percent, one-way streets can hardly be driven, braking can only be done slowly and a greater following distance must be observed. The number of stops increases from an average of 1.8 to 3.3, an increase of 80 percent. In addition, in autonomous car-free traffic, traffic jams are 55 meters long, compared to 114 meters at 50 percent.

Lohmiller stressed that cities need to think about it, because with the advent and spread of new technology, things will not get any better automatically. One solution may be lanes for autonomous cars, but they will not always have enough space and will increase in number, which is not the way to go. The more radical idea of ​​banning man-made vehicles is unlikely to be politically feasible in the foreseeable future.


Simulations suggest that the advent of autonomous vehicles will degrade traffic flows, with only above 50 percent benefiting.

Knoll estimated it would take around € 1 billion to set up a digital twin system on major German motorways. Every 150 and 500 meters, one sensor should be installed that can display traffic in real time. This allows you to plan traffic for the next 60 minutes with great precision, activate traffic systems, optimize the phases of traffic lights, and redirect robotic cars. Moreover, in principle, autonomous vehicles can also be controlled centrally. This allows them to have a uniform speed and high traffic throughput. Knoll pointed out that this is already working in China.

The long-term effects could be even more important: public authorities and road network operators would be able to react more quickly to accidents and congestion, and fewer roads would have to be built to divert similar levels of traffic. This would make much better use of the available transport space. This would mean fewer congestion and costs overall.

In 2019, Deloitte discussed that self-driving taxis could significantly increase downtown car traffic by 2035. And in October 2020, staff at the Vienna University of Technology determined that autonomous cars would increase traffic.


Source: SG.hu Hírmagazin – Autó by sg.hu.

*The article has been translated based on the content of SG.hu Hírmagazin – Autó by sg.hu. If there is any problem regarding the content, copyright, please leave a report below the article. We will try to process as quickly as possible to protect the rights of the author. Thank you very much!

*We just want readers to access information more quickly and easily with other multilingual content, instead of information only available in a certain language.

*We always respect the copyright of the content of the author and always include the original link of the source article.If the author disagrees, just leave the report below the article, the article will be edited or deleted at the request of the author. Thanks very much! Best regards!