Baidu unveils an upgraded autonomous driving platform Apollo 2.5

CiDi-Apollo

Baidu has unveiled Apollo 2.5, the latest iteration of the open autonomous driving platform, which now supports autonomous driving on geo-fenced highways. At a launch event to celebrate Apollo’s one-year anniversary in Beijing, Baidu also announced the Apollo Automotive Cybersecurity Lab to further the advancement of safety in mobility as a long-term vision towards the development of an autonomous driving ecosystem.

Baidu has teamed up with China Automotive Technology and Research Center and China Academy of Information and Communications Technology to launch its automotive cybersecurity lab. The 1,000m2 lab, located in Beijing’s Haidian district, will bring together talent from Baidu’s Intelligent Driving Group with industry and academic experts from universities, such as Tsinghua University and Beijing Institute of Technology, to focus on in-depth research as the foundation for the sustainable development of intelligent driving.

The lab will conduct research into automotive cybersecurity technologies, trends and solutions for in-vehicle systems, car-to-car communications, the controller area network (CAN bus), and sensors. It will explore best security practices in data protection, in-vehicle infotainment, reference software and hardware designs, and countermeasures to fake signals that misguide autonomous driving systems.

Leveraging its capabilities in autonomous driving and cybersecurity, Baidu will begin to work with Apollo partners, including Chery, BAIC’s Beijing Electric Vehicle Co. and China FAW Group, to establish various joint labs in China. In the future, Baidu will launch research projects with universities, research institutes and cybersecurity teams, and will share their technology developments with partners in the Apollo ecosystem.

On Apollo’s one-year anniversary, Baidu announced the latest iteration of the autonomous driving platform – Apollo 2.5 – which will further expand its capabilities into a wider range of applications and scenarios. It implemented a geo-fenced, high-speed autonomous driving scenario, and reduced 90% of the sensor cost to better facilitate autonomous driving development. The latest iteration also includes vision-based perception, real-time relative mapping and high-speed planning and control.

Apollo 2.5 offers more efficient development tools, including a visual debugging tool, a data collector for HD maps and a simulator for cloud-based autonomous driving, to further enhance developer efficacy. In addition, Apollo’s recently-launched data and technology project, ApolloScape, will release new large-scale autonomous driving datasets and next-generation simulation technologies in China.

Apollo has now amassed more than 200,000 lines of code, which have been downloaded and used by over 2,000 developers and partners. Over 9,000 developers on Github, the world’s largest open source community, have recommended Apollo open source software. After its most recent partnership with BYD, China’s leading electric vehicle manufacturer, Apollo has now brought together 100 partners with the open platform. Furthering its ability to advance the development of autonomous driving, this year, Baidu has been granted permits to perform autonomous vehicle road testing in China’s Beijing, Fujian and Chongqing.

Source: Baidu