California Releases 2018 Self-Driving Disengagement Reports
Last week, the California DMV released the most recent disengagement reports for companies testing autonomous vehicles. While many have reported that Waymo and Cruise are leading the pack, with Apple lagging behind, these statistics can be misleading. The reports can give some idea of the progress within individual companies, but it is difficult to compare across companies because of differing interpretations of what constitutes a disengagement and unknown testing conditions. This highlights the need for AV safety performance metrics that better correlate with actual safety and are comparable across a variety of systems. The creation of these metrics is critical for supporting safe AV deployment and garnering public trust in the technology.
While the U.S. scores highly due to the abundance of AV testing and companies located in the country, it is falling significantly behind on legislation
KMPG: U.S. Slipping in AV Readiness
Professional service company KPMG has published its second Autonomous Vehicles Readiness Index, which evaluates which countries are best prepared for autonomous vehicles. Countries are assessed in four main categories: policy and legislation, technology and innovation, infrastructure, and consumer acceptance. The top five countries in this year’s survey were the Netherlands, Singapore, Norway, the United States, and Sweden. While the U.S. scores highly due to the abundance of AV testing and companies located in the country, it is falling significantly behind on legislation (ranking ninth in this particular category). Nationwide legislation will be required for optimal deployment of autonomous technology and reports such as this emphasize that the U.S. will fall behind globally if such legislation is not implemented.
Uber and GM Cruise are making their respective AV ‘visualization’ tools open source
Uber and GM Cruise have both announced that they are putting their AV visualization tools, respectively called Autonomous Visualization System and Worldview, online and making them available for free. AV visualization is used to make raw data from self-driving cars more easily understandable so that operators can have better insight into how their vehicles interact with surrounding objects. Both companies have emphasized that publishing their visualization software will allow people to utilize these tools without having to learn complex graphics techniques. This could make it easier for companies to figure out why their cars made the decisions they did and accelerate AV technological development across the ecosystem.
Audi Cars Tell You How Fast to Go to Catch All Green Lights
Audi is introducing a feature in some models called the Green Light Optimization Speed Advisory system as an expansion of their vehicle-to-infrastructure system. The technology will tell drivers how fast they should drive if they want to only catch green lights. This is currently being pitched by the automaker as a stress-reduction tool for drivers, but it could also be essential to realizing the full potential of future autonomous vehicles. Currently, constant stopping and starting at lights is a source of wasted fuel. Including this technology in AVs could increase the speed and efficiency of traffic in congested areas and reduce system-wide emissions.