The Next Big Thing! Who is “the Real Eye of the Car”? — Part1
In the spring of 2022, as smart driving enters the second half, the two camps start planning quickly. As the race is about to start, more car companies began to focus on “visual-based solutions”.
The Next Big Thing
2021 is recognized by the industry as the first year for intelligent driving to enter the L3. China, the US and Europe are in the stage of accelerating industrial layout, policy support and guidance.
It is estimated that by 2025, the global vehicle shipments with autonomous driving (assistance) functions (including L1 to L5) will be 63.32 million units, and the market space will grow significantly from $17.1 billion in 2020 to $78.1 billion in 2025.
Global Intelligent Driving Development and Future Commercialization?
The leaders in smart driving are mainly in the United States and China. The US has dual first-mover advantages in capital and technology, which can be seen from the development of companies such as Tesla, Waymo, and Mobileye, which was acquired by Intel.
Policy support is also sufficient. For example, the California government has been actively guiding intelligent driving policies and loosening regulations for many years. As the largest market for new energy vehicles, China has unique advantages in terms of new technology policies and data regulations, such as NIO, XPENG, LI and other new forces, autonomous driving companies for mass production of passenger vehicles, and RoboTaxi and RoboTruck which directly oriented to L4.
Additionally, as Europe is the birthplace of the automobile industry, it also has a certain leadership in technology application. This year has seen a number of industry landmark events. For example, Mercedes-Benz, which is scheduled to be delivered in the first half of 2022, will become the world’s first L3 autonomous vehicle to achieve mass delivery and road-legal, capable of driving at speeds of up to 60 km/h.
In addition, Porsche is preparing for an IPO, which would be the largest automotive IPO in European history. It is clear that the purpose of financing is used to support investments in electrification and intelligence.
We can see that autonomous driving is in fierce competition around the world, which also proves that commercialization and large-scale landing have started.
How to Cross “Death Valley” from L2 to L3
Someone has ever said that L2 is a baby of one or two years old who can stumble around a few steps. While L3 is a 3–5 years old kid who can walk on the road, but still need the parents’ help.
The leap from L2 to L3 is a qualitative step, which can be called a “Death Valley” kind of leap.
The responsibility subject of driving changes from human to the car as it crosses from L2 to L3. L3 allows consumers to hand over the decision-making power to automobiles under limited conditions. In this case, the machines judge and make decisions themselves without human instructions and operation. Therefore, the automobile and the manufacturers should be held responsible for possible accidents.
In other words, at L3, the car is far from being merely a vehicle, but an “intelligent robot”. Future products can be mass-produced and replace drivers. They replace humans to make decisions and assume corresponding responsibilities.
Large Amount of Labeled Data Is Needed
The mainstream algorithm model of autonomous driving is mainly based on supervised deep learning. It is an algorithm model that derives the functional relationship between known variables and dependent variables. A large amount of structured labeled data is required to train and tune the model.
On this basis, if you want to make self-driving cars more “intelligent”, and form a closed loop of the business model for self-driving applications that can be replicated in different vertical landing scenarios, the model needs to be supported by massive and high-quality real road data.
In the field of autonomous driving, data annotation scenes usually include changing lanes and overtaking, passing intersections, unprotected left and right turn without traffic light control, and some complex long-tail scenes such as vehicles running red lights, pedestrians crossing the road, and roadsides as well as illegally parked vehicles, etc.
The current artificial intelligence is also called data intelligence. At this stage of development, the more layers of the neural network, the larger amount of labeled data is needed.
For deep learning, data is meaningful only if it is well labeled.
Outsource your data labeling tasks to ByteBridge, you can get the high-quality ML training datasets cheaper and faster!
- Free Trial Without Credit Card: you can get your sample result in a fast turnaround, check the output, and give feedback directly to our project manager.
- 100% Human Validated
- Transparent & Standard Pricing: clear pricing is available(labor cost included)
Why not have a try?