The best open source rookies of 2018On March 21, 2018 by Zander
Over the last year, major vehicle manufacturers have been clamoring to be among the first to market with a safe and reliable autonomous vehicle. Volkswagen, Volvo, BMW, Audi, Tesla, Uber, Google, and even Amazon are pushing boundaries with proprietary innovation and exclusive partnerships. This approach has one drawback: a cluttered arena of distinct approaches to autonomous driving.
Baidu is seeking to clear a path for innovation with Apollo, an open autonomous driving platform and flexible architecture. With its first version launched in July 2017, Apollo enables Tier 1 providers, vehicle makers, and startups to build their own autonomous vehicles without the burden of reinventing the wheel” Organizations can accelerate development by drawing on the collective expertise of Apollo partners and Apollo’s unique simulation engine, which contains tens of thousands of autonomous driving scenarios (ADSs), to rigorously test autonomous driving algorithms. Now, organizations can verify that their algorithms meet basic regression rules before they get to the road, without the burden of deriving the test data from scratch.
With 50 partners for its 1.0 release, and more than 90 after its 2.0 release in January 2018, Apollo is working fervently to extend its integration to additional hardware platforms to accommodate a wide variety of sensor, compute, and modeling technologies. Apollo’s latest release includes functions to improve performance on urban roads, multisensor support, obstacle perception, traffic-light detection, and enhanced security features. Moving into 2018, the Apollo team is focusing on productization requirements to extend innovation to a greater portfolio of businesses, while upholding one of its core tenets: Autonomous driving data belongs to humankind, not solely to the company, and the more we all contribute, the more we will benefit from this project.
For more information, visit apollo.auto.