foreword
Apollo is an autonomous driving platform launched by Baidu. It is a comprehensive autonomous driving solution that provides core functions including perception, decision-making, planning, and control, as well as supporting tools such as maps, positioning, simulation, and data management.
Article Directory
The history of Apollo
Counting from the announcement of the Apollo open plan in April 2017, after 5 years of development, Apollo has released 12 versions, the latest version being Apollo 8.0. During these five years, Apollo has continued to innovate in many ways. The whole development process can be divided into three stages.
The first stage is from the initial closed-field tracking automatic driving to the simple urban road conditions of the 2.0 version. Apollo has established basic automatic driving capabilities and infrastructure.
The second stage is from version 2.5 limited area vision high-speed to version 6.0 education version (EDU), Apollo has accumulated rich scene application capabilities and development experience.
The third phase began with version 7.0. Apollo focused on improving the ease of use of the project, starting from the actual needs of developers, and helping more developers to use Apollo better and faster. Each upgrade expands the capability boundaries of the open autonomous driving platform from multiple dimensions.
In this continuous upgrade process, Apollo continues to add new features, improve performance, and gradually cover a wider range of application scenarios to promote the development and popularization of autonomous driving technology.
New Features of Apollo 8.0
On December 28, Baidu's Apollo open platform officially launched a new upgraded version of the Apollo autonomous driving open platform, Apollo 8.0, for all developers, which further consolidated the ease of use of the platform and made it easier for developers to operate.
The Apollo 8.0 version has been upgraded in the following functions:
package management
Version 8.0 rearranges the dependencies between modules and introduces the concept of software packages to manage and publish each module, optimizes the problems of large downloads, inability to use on-demand, and long compilation times during download and installation, and in this Based on the update, an expansion solution based on application scenarios has been released to allow more developers to use Apollo better and faster.
It is mainly reflected in the following aspects:
- Clearer module structure, lower learning threshold;
- A faster deployment method improves compilation efficiency;
- More convenient expansion scheme, convenient for secondary expansion, verification and practice;
perception frame
In order to help developers better improve the development efficiency of the perception module, Apollo 8.0 provides a complete set of end-to-end autonomous driving perception development process, which has been improved in the four main links of data, model, framework and verification, while iterating Optimize the task process and tools of Lan, help developers quickly realize the development, deployment and verification of autonomous driving perception, and improve the efficiency of perception development.
It is mainly reflected in the following aspects:
- Clear task pipeline, various algorithm plug-ins;
- Brand new model training, easy-to-use deep learning model;
- Efficient model management and convenient model verification;
tool chain
In order to further meet the needs of the autonomous driving development process and improve developers' R&D efficiency, Apollo 8.0 combines the pain points and demands of developers to provide a more comprehensive and easier-to-use tool chain to speed up the developer's R&D process and get started with Apollo better and faster.
It is mainly reflected in the following aspects:
- New and complete PnC tool chain;
- Convenient visualization tool for perception results;
- Dreamview configuration center;
summary
Apollo 8.0 has been comprehensively upgraded from the two important aspects of "new architecture" and "new capabilities", and has been improved based on the actual needs of developers to help developers better and faster familiarize themselves with and use the Apollo open platform. The launch of the Apollo Open Platform 8.0 once again makes the Apollo Open Platform a big step forward in terms of engineering ease of use, lowering the threshold of operation difficulty and operating cost, so that more developers can easily and conveniently get started with the Apollo Open Platform, Into the field of autonomous driving technology.
cloud experience
The cloud lab is based on the real Apollo operating environment, and developers are guided by the experiment manual to experience Apollo in the cloud environment. Here you can start the experimental environment with one click, experience the real autonomous driving scene, and also refer to the detailed experimental manual.
The following list of experiments can be used for your own experience
Let's randomly enter an experiment to see how the quality of the content is?
The experimental content, experimental purpose, and experimental process have all been sorted out for everyone, which is so considerate! Swipe down again, and the experimental phenomenon has also been sorted out for everyone. As the saying goes, we can see the essence through the phenomenon. This is not the same as telling everyone the essence directly, which is great!
Why don't you hurry up and experience it?
package installation
You can refer to the documentation to install it yourself, let’s talk about a few installation commands:
Remember to update related software after installing Ubuntu Linux
sudo apt-get update
sudo apt-get upgrade
Install the Apollo environment management tool and remember to add the apt source
sudo bash -c "echo 'deb https://apollo-pkg-beta.cdn.bcebos.com/neo/beta bionic main' >> /etc/apt/sources.list"
wget -O - https://apollo-pkg-beta.cdn.bcebos.com/neo/beta/key/deb.gpg.key | sudo apt-key add -
sudo apt update
Summarize
The Apollo platform provides a full-stack development environment, enabling developers to conduct autonomous driving-related algorithm research, software development and system integration on the platform. It supports a variety of hardware devices and sensors, such as lidar, camera, millimeter-wave radar, etc., and can be integrated with various types of vehicles.
One of the core modules of the Apollo platform is the Apollo positioning system, which achieves high-precision positioning based on a variety of sensor data. Another important module is the Apollo perception system, which acquires and processes environmental information through various sensors to realize the perception of objects and road conditions around the car. In addition, Apollo also provides a decision system and a planning system for real-time decision-making and path planning.
As a complete open platform, Apollo also provides a wealth of development tools and resources, including simulation environments, datasets, algorithm libraries, and development documents, to help developers develop and test autonomous driving systems more efficiently.
Activity
2023 Spark Training [Perception Special Camp] is created by Apollo developer community preachers. The new perception course is online. Four lessons will take you to get started with the use and development of Perception 2.0!
activity description
Ways of learning
class schedule
Activity Award
ways of registration
Interested friends come to participate quickly!