On Road

The supplier now offers video perception as a modular software package that can be used on various systems on chips so customers can put together assisted and automated driving and parking solutions that best suit their needs.

At the IAA Mobility 2023 show next week, Schaeffler will exhibit the first electric demonstrator and its electrification and steering technology and the plans for Mobileye to provide its SAE Level 4 autonomous driving system.

This week’s best practice for developing ADS safety performance thresholds based on human driving behavior follows last month’s publication for continuous monitoring and improvement after deployment, as its Automated Vehicle Safety Consortium accelerates support for AV advancement.

At the CAR Management Briefing Seminars earlier this month, Ghost Autonomy’s CEO discussed how his company designing for the volume car market with software that runs on commodity hardware designed with universal AI, no HD maps, and fully auto-labeled training data.

The technology, to be integrated by ECARX, is planned to offer eyes-off, point-to-point autonomous driving on highways, as well as eyes-on automated driving for other environments in identified operational design domains.

The card paired with Recogni’s SDK provides greater development and deployment flexibility for any autonomous mobility use case for what it says is the world’s first 1000 TOPS class inference solution.

Its MK6 connected-vehicle solution has achieved CE certification in Europe, paving the way for its implementation in the region for the rollout of intelligent transport systems and an enabler for highly automated driving introduction.

Google’s sibling Waymo is focusing on robo-taxis instead of autonomous trucking.

Innoviz is developing an enhanced solution including the InnovizTwo lidar, a new InnovizCore compute box, and AI-powered software solutions including an industry-first MRM system to enable Level 3 automated driving capabilities to a broader range of future vehicles.

New strategy can help avoid complex models of the self-driving vehicles and their environments, as well as the expensive, time-consuming validation of those models.