With more and more self-driving cars taking to the streets one after another this year, they are truly serving the public, marking that this technology has entered a critical moment before mass production. In order to meet the upcoming first year of commercialization of autonomous driving in 2020 with a better attitude, almost all the entrants are currently seizing the time window period, speeding up the pace of research and development and promoting the listing of products.

On September 17, Huanyu Zhixing, a new technology company focusing on the research and development of networked autonomous driving systems and components, held the "Unmanned Intelligent System Athena Press Conference" in Shanghai and officially released the Athena high-reliability software system. It is understood that Athena mainly runs on the Titan autonomous driving domain controller independently developed by Huanyu Zhixing. The combination of the two marks the official completion of the construction of the open platform of Huanyu Zhixing's software and hardware integration.

As an emerging technology company in the field of autonomous driving, it is reported that Huanyu Zhixing began research and development on autonomous driving as early as 2008. So far, it has established a combination of lidar and camera fusion perception positioning, decision planning, simulation platform, and intelligent networked fleet A series of leading autonomous driving technologies, including hardware and software, cover two different aspects.

In terms of hardware, the latest breakthrough of Universal Smart is the launch of TITAN III autonomous driving domain controller, which is based on NVIDIA's TX2 design, runs the UBUNTU operating system, and supports various hardware interfaces such as USB, CAN, RS232, Ethernet, AD/DA, etc. Access, can simultaneously access 12-channel camera, 6-channel lidar and 5 millimeter-wave radar. Not only that, TITAN III also has the advantages of small size, chip-level algorithm integration, high-reliability vehicle-level standards, and low power consumption.

TITAN III integrates all the components required for the L4 automatic driving function on a single board, including perception fusion computing, high-precision positioning, vehicle control, and cloud interactive communication. No other external equipment is required, only the camera and lidar need to be connected , Millimeter wave radar, you can achieve a variety of intelligent functions such as high-precision maps, ACC, LDW, LKA, FCW, LC, VUI, PDS, AEB, one-click overtaking.

At present, a large number of facts prove that the car will no longer be a cold machine in the future, but will be empowered by various software to continuously expand its connotation and extension, and obtain more functions than traditional travel. By integrating Athena with TITAN's autonomous driving domain controller, Huanyu Zhixing has realized high-definition map acquisition, real-time three-dimensional fusion perception and positioning, data storage and high-speed Ethernet communication, as well as different forms of autonomous driving functions for bicycles, fleets and vehicles, It makes it possible for car companies to quickly verify autonomous driving technology and shorten the mass production time.

In order to better prove the software capabilities of Universal Smart, at the press conference, the relevant person in charge of Universal Smart also launched their latest developments in the field of AI, 3D perception, multi-sensor fusion positioning, and dynamic planning based on intelligent network technology. Explain and make a one-click video demonstration of the car call and return, and a remote demonstration of the 5G networked automatic driving function.

According to Dr. Cao Jing, Huanyu Zhixing, thanks to the development of artificial intelligence technology, some new changes have also occurred in the field of autonomous driving, such as the change from the visual perception around 2D images to the current 3D visual perception. In this regard, Huanyu Zhixing has made many breakthroughs, including the physical description of three-dimensional information and deep learning directly on the point cloud. In automated driving testing, it has become possible to replace more actual road tests with virtual simulation technology, especially in the generation of strategies and plans. Huanyu Zhixing makes extensive use of cutting-edge simulation technologies, including software-in-the-loop and hardware In the ring, the vehicle is in the ring, etc., making the result more reliable and stable.

Academician Zhang Jianwei, co-founder of Huanyu Zhixing, brought everyone the challenges of artificial intelligence next. He pointed out that the current artificial intelligence is still weak artificial intelligence, not easy to make small mistakes but easy to make big mistakes. How to use human-like multi-mode It is one of the important directions of basic scientific research to achieve strong artificial intelligence with dynamic information. Academician Zhang also pointed out that with artificial intelligence for autonomous driving, the future should focus on considering the direction of multi-modal information fusion of more humans in addition to vision and lidar. At the same time, it can learn from the current aspects of reinforcement learning, continuous learning, and transfer learning. Achievements to enhance the learning ability of autonomous vehicles. Academician Zhang also mentioned the need to solve the "three outlooks" problem of autonomous driving, especially the moral outlook, and multi-robot games for large-scale group intelligence. The research of these key issues and the research and development of related technologies are also the direction of academic cooperation between Academician Zhang's German team and Huanyu Zhixing Institute.

According to Li Ming, CEO of Huanyu Zhixing, the current program of software and hardware integration of Huanyu Zhixing is nearing completion. The next phase will focus on how to further improve the reliability of the open platform on the basis of the existing basis, and let this The solution is commercialized. "In November, at Nvidia's GTC China 2018 conference, we will release specific commercialization paths, including automatic driving application scenarios for different models such as heavy trucks and passenger cars. At the same time, we also have some The cooperation project is in progress, mainly for the development and testing of L4 level autonomous vehicles." Li Ming said.