Newbies on the Road: A Quick Guide to Getting Started in the Autonomous Driving Industry

1. Development of autonomous driving technology

1.1 The industrial revolution drives the development of autonomous driving technology

With the improvement of infrastructure, the development of manufacturing industry, and the advancement of information technology, the industrial revolution has brought many conveniences to mankind.

**But human beings’ pursuit of a better life never stops. **People who don’t have a car want to drive a car; people who have a car want to have a driver.

What? Are drivers too expensive? Can the car be driven by itself?

The industrial revolution brought about the rapid development of information technology and provided necessary technical support for the realization of autonomous driving technology. For example, advanced sensor and computer vision technologies enable precise navigation and autonomous driving of vehicles.

1.2 Imaginary future: self-driving cars in science fiction works

When I was a child, I often sat in front of the TV and watched those imaginative science fiction works. Among them, self-driving cars have left a deep impression on me.

The city streets of the future will no longer be noisy or congested. Shining cars drive freely on the road. They have no drivers, no steering wheels, or even accelerators and brakes.

These cars are like smart elves, traveling freely according to their own wishes and destinations.

1.3 Germination and attempts of autonomous driving technology

The germination of autonomous driving technology can be traced back to the early 20th century, when people began to try to use various technologies to realize autonomous driving of cars.

The first car thought to have self-driving capabilities was the car known as the "American Miracle," which debuted in 1925. This car realizes autonomous driving through radio control. Although it only implements simple functions such as starting the engine and sounding the horn, it is one of the early attempts at autonomous driving technology. Similar to the current "toy remote control car".

In the following decades, as technology continued to advance, autonomous driving technology gradually developed. People are beginning to use radar, lidar, cameras, sensors and other devices to perceive their surrounding environment, and analyze and process this information through computer vision, artificial intelligence and other technologies. The development of these technologies makes it possible to realize autonomous driving.

1.4 A hundred flowers contend: Autonomous driving technology giants and startups advance side by side

At the beginning of the 21st century, with the rapid development of the Internet and computer technology, autonomous driving technology ushered in new development opportunities.

In China, as early as 2013, Baidu established a research and development team and began to explore autonomous driving technology.

On December 9, Apollo open platform 9.0 was released. The core theme still revolves around development friendliness, optimization of engineering framework, and improvement of software and application scenario capabilities. At the press conference, Zhang Liang, general manager of Baidu Autonomous Driving Platform Ecology Department, said: In the past six years or so, the entire Apollo open platform has accumulated about 100,000 developers. In the past year, the number of our developers has soared to 160,000, and the growth rate is more than three times the average growth rate over the years.

The development of autonomous driving technology is in a period of unprecedented historical opportunities.

2. How do individual developers get started with autonomous driving?

For individual developers, whether it is out of interest or to grasp the trend of the times, autonomous driving is a very good field.

In addition to developers learning basic knowledge in the Apollo developer community, Apollo Open Platform 9.0 also brings many new changes and upgrades to the engineering framework, algorithm capabilities, tools and documentation, making it easier for developers to get started, use it flexibly, Scenario applications are more versatile and easy to expand.

2.1 Flexible and easy to use, engineering framework upgrade

• The early traditional package management model: using full source code download and compilation, the installation process is cumbersome.
• Apollo9.0 adopts a modular management method, split into smaller granularities, and can be installed flexibly on demand.

Significantly reduce secondary development costs: it only takes 1 day to complete the application scenario demo construction, quickly locate the parameters that need to be adjusted, and the configuration and adjustment efficiency is doubled, and function expansion can be achieved through the required template interface, and the code learning cost is reduced by 90%. Volume is reduced by 50%.

2.2 Rich and comprehensively enhanced algorithm capabilities

The algorithm capabilities are comprehensively enhanced: the main model is more advanced, has better effects, and the accuracy of call rate greatly exceeds the old model;

2.3 Comprehensive upgrade of tool chain to DreamView+

The debugging process is simpler, the window layout is more flexible, and resource access is more convenient.

• DreamView+: Free drag-and-drop panel, highly customized to meet different usage habits and needs.
• DreamView+: Cloud-integrated resource center, unified resource management, DV side can be accessed with one click anytime and anywhere.
• DreamView+: Pattern-oriented multi-scene workflow, the process is more concise, highly optimized, seamless and intuitive.

2.4 Convenient and enriched, document platform reconstruction

The document platform has been comprehensively upgraded and reconstructed to reduce the cost of learning and use, making operation more convenient, reading smoother, and the content more substantial.

Resource packages, APIs, and process guides are all documented.

2.5 Large-scale implementation of application scenarios

In terms of implementation, Baidu Apollo has also reached many cooperations with Suzhou Jinlong, a domestic passenger and commercial vehicle head manufacturing company.

Suzhou Jinlong has actively invested in the research and development of autonomous driving technologies and products, and has invested in the research and development of a series of autonomous buses and low-speed unmanned operating vehicles .

Among them, low-speed unmanned operating vehicles include three categories : unmanned delivery, unmanned sales, and unmanned cleaning .

Baidu Apollo also provides assistance to Suzhou Golden Dragon in terms of talent training, technology improvement, business traction and other aspects.

2.6 University production, teaching, research and application, customized training of industry talents

• Apollo Studio developer community: Build a one-stop learning practice community for courses, experiments, and events.

• Apollo EDU upgraded version released, undergraduate and vocational education customized training cooperation.

In terms of competitions : 300+ registered schools; 1,500+ participating teams; 5,000+ participants; the largest registration scale for autonomous driving events in the country.

In terms of courses : 3.8 million people studied online; currently the largest online training for autonomous driving in China

In terms of institutions : For undergraduates and vocational colleges, we provide customized and three-dimensional school-enterprise cooperation talent training solutions, and carry out cooperation in all aspects such as teaching, teachers, innovation, practical training, and cooperation support; a total of autonomous driving-related courses have been opened across the country. Professional science and engineering colleges cover 70%+; 985 and 211 colleges cover 68%

3. The future of autonomous driving

The future development of autonomous driving will bring about many changes, including improving traffic efficiency, reducing accidents, optimizing travel experience, etc.

Technological innovation is key, and with advances in sensor technology, computing power and artificial intelligence algorithms, autonomous driving will become more reliable and safer.

With the release of Apollo Open Platform 9.0, I can see that Baidu Apollo, as the leading domestic autonomous driving technology company, has made a lot of progress. We look forward to the Apollo open platform continuing to maintain technological innovation and promoting the commercial application of autonomous driving.

In the future, with the popularization of autonomous driving technology, urban transportation will become more intelligent and efficient, bringing more convenience and safety to mankind.