The first self-driving bus in Yongchuan, China on 12th April 2021
Automatic driving level: L4
Power: new energy
Function: Have the capacity to pull at the bus stop and cope with the bus stop scenario and more complex urban road conditions.
Sensing equipment: The bus has 4 lidars, 2-millimeter wave radars, 7 monocular cameras, 360-degree design without dead angle. More comprehensive sensing enables vehicles to monitor all kinds of information in the process of driving.
At 12 noon on 12th April 2021, a red Baidu driverless bus slowly drove into the bus stop of Yongchuan gymnasium, and the citizens waiting at the bus stop began to get on the bus in rows. The seemingly ordinary scene conveys the message that the first self-driving bus in China has been put into operation, marking a major breakthrough in the commercialization of self-driving in China.
Two-way mileage: 10km
You can swipe your card when you get on the bus
Three self-driving buses were put into operation in the first batch, with a speed of 40–60 kilometers per hour, running a two-way mileage of nearly 10 km. Like other buses, passengers can swipe your card when they get on.
Mr.Xiong, general manager of Baidu’s intelligent transportation business in Chongqing, explained that as an autonomous bus on open road operation, the intelligent perception system collects high-precision data of all traffic participants on the road.
Combining the current pedestrians and cars on the road with v2x((Vehicle to Everything), the self-driving bus can realize L4 level driving, it can also receive real-time information from the roadside as well as reflect the blind areas that cannot be seen from the driver’s perspective. The pre-warning of the traffic light changes and waiting time permit the vehicle to make decisions and plan in advance, improving the traffic efficiency, fully meeting the needs of the normal operation of public transportation.
The car runs smoothly
At 12:00, the journalist was on the first self-driving bus for a ride experience. There is a safety officer in the car. Although he sits in the cab, the whole driving process is basically controlled by AI, and the safety officer doesn’t even need to hold the steering wheel.
The seats of the car are very comfortable. After getting on the bus, you can swipe your card to pay the fee. When you meet a red light or a pedestrian crossing the road, the car will stop slowly. When you go uphill or make a turn, the car will adjust its speed automatically. The whole operation is relatively smooth.
An aunt who got on the bus at Yongchuan gymnasium station said: “I didn’t expect that the car would move so smoothly, it didn’t shake at all, and there was no sudden braking. The seats were very comfortable. Next time I have a chance, I will get on the self-driving bus.”
There are two inside large screens in front of the car, one is for the real-time driving distance and the line outside the car, and the other is for the real-time monitoring of the cab.
During the whole driving process, the safety officer basically did not drive the vehicle and even did not need to hold the steering wheel.
It is said that according to the relevant standards, the L4 level automatic bus needs to be equipped with safety officers, who should sit in the driver’s cab only to solve the emergency situation, not to drive the vehicle.
There are five levels of automatic driving, from L1 to L5. The L5 level has the real meaning of “driverless”, there is neither a safety officer nor a steering wheel, however, the L5 level landing operation still takes time.
According to reports, as the first city in China to start the commercial operation of automatic driving, Chongqing citizens can make an appointment through app ports such as Apollo go, Baidu map, and Yongchuan service commune to experience automatic driving.
The data annotation is supposed to make machines understand the world. In auto autonomous driving, the annotation scenarios usually include changing lanes to overtake cars, passing intersections, unprotected left turns and right turns without traffic light control, and some complex long-tail scenarios such as vehicles running red lights, pedestrians crossing the road, vehicles parked illegally on the roadside, and so on.
Several data annotation tools commonly used in auto autonomous driving
In the field of self-driving, the annotation tools commonly used include 2D boxing, 3D cube, lane line, polygon, semantic segmentation, point Cloud, and so on.
- Dealing with complex tasks, the task is automatically transformed into tiny components to minimize human errors
- The real-time QA and QC are integrated into the labeling workflow as the consensus mechanism is introduced to ensure accuracy
- Consensus — Assign the same task to several workers, and the correct answer is the one that comes back from the majority output
- All work results are completely screened and inspected by machines and the human workforce
Thomas C. Redman sums up the current data quality challenge in this way: “Increasingly complex problems demand not just more data, but more diverse, comprehensive data. And with this comes more quality problems.”
Bytebridge is dedicated to empowering the machine learning revolution with no bias training data.
We can provide personalized annotation tools and services according to customer requirements.
If you need data labeling and collection services, please have a look at bytebridge.io, the clear pricing is available.
Please feel free to contact us: email@example.com
Chongqing Daily Journal
1 Data Annotation Service — How an Automated Data Labeling Platform Fuels Autonomous Vehicles Industry?
2 How Auto-Driving Achieved through Machine Learning?
3 Labeling Service Case Study — Video Annotation — License Plate Recognition
4 What is Semantic Segmentation, Instance Segmentation, Panoramic segmentation?
5 High-Quality Training Data for Autonomous Cars in 2021