Mitsubishi Electric Corporation announced yesterday (March 16, 2017) that it is based on artificial intelligence (AI) technology and its mobile mapping system (MMS), Two new technologies have been developed - automated mapping technology and technology for extraction of transitions in mapping landscape. The above technology can provide users with static information of roads and surrounding objects, and apply it to the drawing of dynamic maps. As a pioneer in the industry, Mitsubishi Electric aims to shorten the drawing time of maps and continuously push the latest dynamic information (such as traffic lights and surrounding vehicle information) to ensure the safety and high accuracy of automatic driving. CeBIT 2017 will be held on March, 2017, at which Mitsubishi Electric will exhibit these two new technologies for the first time
automatic mapping technology
with the help of artificial intelligence (AI) technology, the system can quickly create precise and accurate 3D maps, and extract the required information such as road signs and traffic signs from laser point cloud data (also known as "point cloud data", laser point clouds) and camera data. The above data are measured and collected by mobile mapping system. The system will provide users with 3D location information of roads and roadside buildings, and its absolute precision is equal to or less than 10 cm. When the vehicle is driving, the on-board laser scanner, camera and GPS antenna will form a cooperative system to jointly collect the required location information. AI technology also improves the accuracy of data. It only extracts and identifies the necessary information, making the speed of map creation 10 times faster than the manual creation method commonly used in the industry, and its cost is much cheaper than the conventional way
feature point change extraction technology
at present, Mitsubishi Electric combines several extraction technologies to create dynamic maps earlier, so as to improve the efficiency of map update and shorten the time required for map maintenance. This technology can automatically extract the characteristic points of the past data, and compare the latest laser point cloud data measured by the tear resistance experiment of plastic film (ellemendorf method) of the mobile mapping system. This technology can identify the differences between the two and extract the information of the variable feature points. With the help of this advanced technology, the maintenance time of dynamic map and the upgrade time of high-precision 3D map are greatly shortened. The system will only extract the changed feature points for map update, and it is not necessary to update the whole map every time
in the future, Mitsubishi Electric plans to sell its software using the above technology to map publishers. It is expected that Mitsubishi Electric will reach a software sale deal with dynamic map planning corporation in October this year, which will use the software to draw Japan's Highway Map and present high-precision 3D maps to users
background introduction
it is estimated that in addition to this, these areas also need to be often coated with anti rust oil. In, Japan's automatic driving level will be improved from advanced driving assistance system (ADAS) to level 3 automatic driving at the 2014 China International Adhesives and Sealants Exhibition held at the Shanghai WorldExpo Exhibition Hall on September 25. The technical demand for auto drive system will further increase. Because the auto drive system needs to combine the sensors in the car with dynamic maps, the biggest challenge is how to provide users with the latest version of map information. Mitsubishi Electric's two new technologies can quickly and efficiently create and update 3D maps with high accuracy. In the future, Mitsubishi Electric will continue to work to shorten the creation time and maintenance time of maps and realize the dynamic update of autonomous driving maps. 1025 achievements of new material industry
Copyright © 2011 JIN SHI