Skip to main content

Toyota demos HD maps for Tokyo AVs

Toyota's Research Institute-Advanced Development has demonstrated high definition (HD) maps for autonomous vehicles (AVs) for surface roads in Japan with a relative accuracy of less than 50cm.
By Ben Spencer March 31, 2020 Read time: 2 mins
Toyota demos HD maps for AVs for surface roads in Japan (© Rui Baião | Dreamstime.com)

Toyota is hoping to use the results to shorten the delay in updating HD maps for AVs, expand map coverage and reduce maintenance costs.
 
The company collaborated with space infrastructure company Maxar Technologies and IT services provider NTT Data to build the maps using commercial satellite imagery.
 
The demonstration allowed the partners to automatically extract the required map information by analysing, removing and correcting non-map image pixels such as cars and shadows due to the inclination of buildings in satellite imagery. Maps were created within the 23 wards of Tokyo as well as six unnamed cities in the US and Europe.

A collaboration with road intelligence firm Carmera showcased how to make maps using consumer-grade vehicle cameras. Both parties used dashcam drive recorders to detect and place road features such as lane markings, traffic signals and signs within the 23 wards of Tokyo and two unnamed cities in the US state of Michigan.

In a separate initiative, Toyota showed how lane markings necessary for AVs could be updated in near real-time on TomTom's HD map. This was achieved by converting vehicle data into TomTom's cloud-based transactional mapmaking platform, Toyota says.
 
Additionally, Toyota worked with Here Technologies to correct the positional errors in its vehicle data. Here created surface road maps and ingested data into its platform and automatically generated HD maps including lane level information, the company adds.

 

For more information on companies in this article

Related Content

  • Cloud computing technology benefits GIS
    July 17, 2012
    Geographic Information Systems are a relatively late adopter of cloud computing,but the benefits of host services for geospatial data and analysis are becoming clear. Jason Barnes reports Both the concept and the reality of cloud computing have been around for some time. More and more industry sectors are entrusting external service providers with the provision of their computing services via the internet. However, the Geographic Information System (GIS) industry has been slow to embrace the trend. This is
  • Advanced ITS truck screening aids border control
    March 14, 2012
    State-of-the-art ITS technologies are being deployed for tracking of commercial vehicles at the US-Mexico border in Arizona, reports Pete Goldin. The border between the US and Mexico may be the epitome of America's wild west, but this remote desert frontier is being tamed by the Arizona Department of Transportation (ADOT) with a state-of-the-art ITS system. A comprehensive port-of-entry (POE) screening system is being deployed at the Mariposa Port of Entry – one of the busiest land ports in the nation – at
  • Advanced ITS truck screening aids border control
    March 14, 2012
    State-of-the-art ITS technologies are being deployed for tracking of commercial vehicles at the US-Mexico border in Arizona, reports Pete Goldin. The border between the US and Mexico may be the epitome of America's wild west, but this remote desert frontier is being tamed by the Arizona Department of Transportation (ADOT) with a state-of-the-art ITS system. A comprehensive port-of-entry (POE) screening system is being deployed at the Mariposa Port of Entry – one of the busiest land ports in the nation – at
  • Targeted roadside advertising project uses deep learning to analyse traffic volumes
    June 22, 2016
    A targeted roadside advertising project for digital signage using big data and deep learning just launched in Tokyo, Japan, by US smart data storage company Cloudian will focus on vehicle recognition and the ability to present relevant display ads by vehicle make and model. Together with Dentsu, Smart Insight Corporation, and QCT (Quanta Cloud Technology) Japan, and with support from Intel Japan, the project will conduct, at its first stage, deep learning analysis – artificial intelligence (AI) for recog