IT House reported on December 19 that Baidu officially launched a new upgraded version of the Apollo open platform today - Apollo open platform 9.0.
Zhang Liang, general manager of Baidu Autonomous Driving Platform Ecology Department, summarized the main highlights of this upgrade: "Apollo Open Platform 9.0 has achieved a comprehensive upgrade of in terms of engineering, algorithms and tools, and the universal layer can empower the scale of a variety of application scenarios. The overall operation is more flexible and easy to use, and the usage scenarios are universal and easy to expand. While greatly improving development efficiency, it can help more developers quickly build their own autonomous driving systems." In terms of the
engineering framework, in order to make developers more To flexibly assemble autonomous driving applications and make secondary development more convenient, Apollo Open Platform 9.0 comprehensively upgrades package management and splits modules into smaller software packages according to the granularity of functions. Developers can more conveniently customize their own software according to their needs. You can choose to use it according to your needs. It also provides a wealth of functional components and plug-ins, and the functional extensions have been improved and optimized. Based on this,
has unified the scheduling interface, and developers can complete the scene demo construction in as soon as 1 day. The simplified parameter adjustment method doubles the parameter adjustment efficiency. The new plug-in mechanism reduces the cost of code learning by 90% and reduces the amount of code. 50%, greatly improving Apollo’s secondary development capabilities. Apollo open platform 9.0 also adapts to the ARM architecture for the first time. In terms of
algorithm, Apollo Open Platform 9.0 has optimized the perception algorithm based on the previous one. Lidar detection uses the relatively new CenterPoint model, and visually uses Yolo X + Yolo 3D model . Millions of levels are poured into these two models. Data training, recall rate and precision rate are said to have been "significantly improved".
Apollo open platform 9.0 also provides incremental training. supports independent completion of model training , which can improve the detection capabilities of special targets and special scenes while maintaining the original detection capabilities of the model, thereby easily improving customization at a lower cost. Scene detection effect. In addition, the new platform fully supports 4D millimeter wave radar , which enhances obstacle detection and extreme weather scene safety. In terms of
tools, in addition to adding new functions based on the original tools, Apollo open platform 9.0 also adds new tools including high-precision map mapping, sensor calibration and integration. The newly upgraded Dreamview + in Apollo Open Platform 9.0 has been comprehensively improved in multiple aspects such as multi-scenario use, free layout, and data resources. The debugging process is simpler, the window layout is more flexible, and resource access is more convenient.
In addition, Apollo Open Platform 9.0 also reconstructs the document platform to achieve more convenient operation, smoother reading, and more substantial content, effectively reducing developers' learning and usage costs.
At this press conference, Dr. Dong Zhaozhi, Executive Vice President of Kaiwo Group and Dean of the Academia Sinica, Qiang Xiaowen, Secretary of the Party Committee, Deputy Headquarters and Dean of the Product Research and Development Institute of the Commercial Research Headquarters of Dongfeng Motor Co., Ltd., King Long United Automotive Industry ( Dr. Liu Mingchun, deputy director of the Forward Technology Research Institute of Suzhou Co., Ltd., and Dr. Liu Junchuan, vice president of Tianzhun Technology Co., Ltd., were also present.
Baidu said that Apollo Open Platform 9.0 has powerful scene-universal capabilities, and is easier to use and expand; adaptation steps are reduced by 40%, code reading is reduced by 90%, and code debugging is reduced by 80% , which can achieve "opening" "Ready-to-use", the self-driving vehicle closed loop can be completed within a week. At the same time, Apollo open platform 9.0 covers more scenarios, and the closed-loop operation of the scenario application business system can be completed in one month. On this basis, the sensor calibration and map creation cycles have been shortened to hours, significantly shortening the implementation time and making it more efficient.
At the same time, Apollo open platform 9.0 has lower hardware cost and richer selection of . The hardware selection supports 3Lidar+4Camera. The camera supports more than 4 manufacturers, upgraded from USB3.0 to GMSL, and the lidar adds 32 lines and 64 lines. Lines and other multi-brand and multi-model equipment, positioning equipment has added support for more than 3 mainstream brands of equipment, which can comprehensively assist the implementation of large-scale applications in different scenarios.
IT Home learned from the press conference that currently, the Apollo open platform has brought together more than 160,000 developers from more than 170 countries and regions around the world. has reconstructed 120,000 lines of code during the development process of Apollo open platform 8.0 to 9.0. The new Added 200,000 lines of code .
