Toyota: Creating HD maps for automated driving on general public roads

High-definition maps generated from satellite imagery in joint field testing with Maxar Technologies and NTT Data (Source: TRI-AD, the same applies below)

  This report covers Toyota's ongoing plans to generate high-definition (HD) maps for autonomous driving on general public roads from cameras, drive recorders, satellite imagery, and other devices mounted on ordinary vehicles.

  TRI-AD (Toyota Research Institute Advanced Development), which was established in March 2018 by Toyota, Denso, and Aisin Seiki, is responsible for the actual work. TRI-AD has developed the Automated Mapping Platform (AMP) open software platform. In AMP, sensor data of vehicles equipped with applications and software created by participating developers is provided to the AMP. In return, developers have easy, open and sustainable access to HD maps generated by the AMP, regardless of vehicle type.

  TRI-AD has conducted two demonstration experiments during the past year. The first involved partnering with CARMERA to automatically generate HD maps based on the video data of cameras and drive recorders installed in ordinary vehicles. The second case was to create HD maps using satellite imagery in collaboration with Maxar Technologies and NTT Data. In March 2020, the company announced that it had successfully completed two proof of concept (PoC) projects and obtained results at an effective level for generating HD maps required for automated driving. At the same time, TRI-AD also conducted field testing with the European map giants TomTom and HERE Technologies, and succeeded in generating and updating the HD maps required for autonomous driving on general public roads on the platforms of both companies. TRI-AD will continue to collaborate with its partner companies to further improve the accuracy of HD maps.

 In March 2020, TRI-AD and Dynamic Map Platform Co., Ltd. (DMP), which provides HD maps for automated driving, agreed to jointly conduct new field testing from April 2020. The PoC aims to demonstrate the efficient update of DMP's HD maps by using TRI-AD's AMP open software platform to detect and process changes in lane-level road features based on images and other data collected from vehicle sensors.

Related reports:
Hitachi: Automatic generation of maps for autonomous driving from navigation system maps（Mar. 2020）
(Hitachi's plan also aims to generate maps for autonomous driving on general public roads at low cost, which is the same as Toyota's plan covered in this report. However, Hitachi's plan is based on navigation maps. And, it differs from Toyota's plan in that it does not necessarily require the same level of accuracy as the HD maps already in place for expressways, and that it incorporates the driving know-how of veteran drivers in areas where driving is difficult into the database it has created.)

Automotive World 2020: Robo-taxis, LiDAR, High-definition 3D map（Feb. 2020）
Nissan and Honda Autonomous Driving: Hands-Off from Highway to Downtown（Feb. 2020）
2019 Year in Review: Reality, Trends and Highlights of AV Industry（Dec. 2019）
JSAE 2019 in Yokohama: Connected technology and services for an autonomous driving society（Jul. 2019）
Automotive World 2019: Practical application of autonomous driving technologies（Feb. 2019）
Dynamic Map to generate 3D maps for Japan's expressways by 2018（Feb. 2017）