How to solve traffic congestion? This time, Musk and Robin Lee thought together

Per reporter: Sun Lei Per editor: Pei Jianru

A few days ago, Musk commented on his social media: "Self-driving cars will amplify traffic to the point of almost madness, because you can't feel the pain of driving yourself." In Musk's view, solving the traffic problem is a very difficult task, calling it the "ultimate boss battle", and even the most powerful humans in the world cannot overcome the heavy traffic.

In this regard, Baidu chairman and CEO Robin Li believes that intelligent transportation based on vehicle-road collaboration will be able to improve traffic efficiency by 15% to 30%, and within 5 years, China's first-tier cities will no longer need to "restrict purchases" or "restrict traffic"; within 10 years, relying on the improvement of traffic efficiency, congestion problems can be basically solved.

Musk's idea: congestion on the ground and underground

In fact, Musk's remarks are not unfounded. In 2019, scott Le Vine research team at Imperial College London in the United Kingdom conducted experiments on the impact of autonomous driving on traffic congestion in 16 different road conditions in 4 cities. The results show that if the proportion of self-driving cars on the road is 25%, it will lead to worsening traffic conditions.

It is worth noting that the proportion of autonomous vehicles in the above experimental roads is 25%, which simulates the initial stages of the popularization of autonomous driving technology. The final results of the experiment are in line with Musk's statement that "in the early days of the popularization of self-driving technology, traffic congestion will be exacerbated."

Back in 2016, Musk complained on social media during a traffic jam and came up with a "solution." At that time, Musk said, "The traffic jam is driving me crazy, I'm going to build a tunnel, I'm going to start digging now..." In 2016, Musk founded the Boring Company, whose main business is to dig tunnels.

Image source: Visual China

According to Musk's idea, he would build a tunnel underground, and the vehicles entering the tunnel would operate mainly through an electrified platform that combines autonomous driving and is equipped with unified computer monitoring. The vehicle only needs to be parked on the platform, after which the platform will descend to the tunnel floor and carry the vehicle at high speed. It is reported that the operating speed of the platform can reach up to 200 km / h.

However, there is a big gap between ideals and reality. In February this year, the Las Vegas Convention Center Loop, a subsidiary of Boring Company, also ushered in a "crisis". The underground tunnel, which boasts "a complete solution to traffic jams," was stuck in traffic during CES 2022. A Tesla owner who was stuck in the tunnel said on social media that CES's huge traffic overwhelmed the tunnel and caused blockages.

It is understood that boring company's Las Vegas Convention Center Loop Tunnel is not the final version of Musk's plan. There is currently no envisaged electrification platform in the tunnel, and the operating speed is less than 200 km / h. Musk said when the tunnel was opened that the current V2X technology is not mature enough, and the information linkage between vehicles cannot be solved, so it is temporarily impossible to use the flatbed car of automatic driving in the tunnel.

The underground tunnel plan did not go well, and Musk started the idea of traffic lights again. His idea was: "Maybe Tesla should make an AI visual device for traffic lights to maximize throughput based on actual traffic and traffic conditions." ”

Smart cars still have to cooperate with "smart roads"

Musk's concept can't help but remind people of the vehicle-road collaboration scheme that many Chinese autonomous driving companies have been deeply engaged in for many years. And netizens also commented on Musk's social media, "China has long had it, this is not a rare thing" "The road signal lights and vehicle information are synchronized, the vehicles are based on the changes of traffic lights to achieve a timely start, and the coordinates are Beijing Yizhuang."

Vehicle-road collaboration technology uses wireless communication and vehicle networking and other technologies to realize dynamic real-time information exchange between vehicles and vehicles, vehicles and roads, and vehicles and people in an all-round way, so as to achieve collaborative management of vehicles and roads, further promote the effective coordination of people, vehicles and roads, and ensure traffic safety and improve operational efficiency.

In 2021, the high-level autonomous driving demonstration area in Yizhuang, Beijing, will complete the construction of intelligent network infrastructure with a total length of 12.1 kilometers of urban roads and 10 kilometers of high-speed roads. This demonstration area integrates perception and computing equipment such as electronic police, radar, fisheye camera and edge computing, which realizes accurate monitoring of intersections and accurate disposal of signal lights to achieve smooth travel for citizens in the demonstration area.

At present, Baidu and other companies have carried out the test and verification of the collaborative scale of landing vehicles and roads in Yizhuang. Baidu chairman and CEO Robin Li believes that intelligent transportation based on vehicle-road collaboration will be able to improve traffic efficiency by 15% to 30%, and within 5 years, China's first-tier cities will no longer need to "restrict purchases" or "restrict traffic"; within 10 years, relying on the improvement of traffic efficiency, basically the congestion problem can be solved.

However, whether it is Musk or Robin Li, both advocate vehicle-road coordination and also pay great attention to bicycle intelligent automatic driving. Tesla is currently optimizing its fully autonomous driving (FSD) beta software, while Baidu provides intelligent driver assistance solutions for automakers on the one hand, and it also iterates autonomous driving technology through radish fast running.

Some people believe that bicycle intelligence may soon allow vehicles to achieve more than 99% of automatic driving capabilities, but for the boundary situation after 99.9999% needs to be solved by vehicle-road coordination. Creating a car that is smart enough seems to be the essence of the bicycle intelligent route, but it is also one of the basic needs to ensure the smooth implementation of the vehicle-road coordination plan.

"Mainland China takes the route of 'bicycle intelligence + vehicle-road coordination', that is, in the absence of roadside support, it relies on bicycle intelligence to achieve limited automatic and assisted driving; with the support of the roadside, it realizes automatic and assisted driving in more scenarios." Chen Xiaoming, deputy general manager of Hisense's intelligent transportation division, believes that due to the complexity of traffic travel, although there is a difference between bicycle intelligence and vehicle-road coordination in the current stage, there is no fundamental contradiction, and the coordinated development of the two aspects is expected to solve traffic congestion.

Daily economic news