news.voyage.autoVoyage

Get started Open in app 2.1K Followers About Follow Sign in Get started Follow 2.1K Followers About Get started Open in app Oliver Cameron · Updated Dec 30, 2020 Introducing Commander A state-of-the-art self-driving A.I. to power a robust robotaxi service Meet Commander, the Brain of our Robotaxi The human brain has evolved over hundreds of thousands of years to the point where a student driver can learn to drive a car with only a few hundred miles of experience. From a student driver’s first mile of driving, they harness the powerful perception, prediction, and planning system built into every human brain that makes us remarkably capable of driving a car. However, humans have emotions, and when it comes to driving, those emotions—being tired, distracted, angry, or under the influence—can cause fatal accidents. At , we are not building general intelligence—flaws and all—instead, we are building a brain with a single responsibility: navigate a car from point A to B safer than a human. A brain that cannot get angry, nor tired, nor answer questions nor read books, but that drives a car with superhuman safety and accuracy. Like any brain, it gets better and better at driving every time it drives. This brain is a self-driving A.I. that we call Commander, which, when combined with Shield , Telessist , and the G3 forms a powerful self-driving technology. … Read more · 6 min read 91 1 Oliver Cameron · Dec 23, 2020 Teaching a Self-Driving A.I. To Make Human-Like Decisions A powerful technique to deliver safe, human-like driving decisions A Safe, Intelligent Self-Driving A.I. The key to commercializing self-driving A.I. is to marry safety and intelligence. A safe self-driving A.I. can be validated—knowing what it can and cannot handle. An intelligent self-driving A.I. gets better over time—learning to make increasingly complex driving decisions without getting tripped up. In essence, the combination of the two delivers a safe ride that feels smooth (i.e., not robotic) and gets you to your destination on-time (i.e., doesn’t get stuck). We’ve previously applied this methodology to our computer vision technology, with safety-focused perception algorithms like Sonic and intelligence-boosting research in Active Learning . Active Learning and Why Not All Data is Created Equal How we can build better machine learning models by using less — but more carefully curated — data news.voyage.auto We are now excited to share our work applying this methodology to behavior planning (the module responsible for making key driving decisions). Recent versions of Commander —our self-driving A.I.—feature a new behavior planner that’s delivering human-like decision-making, but without other methods’ gargantuan data requirements. Fueling this method is a new form of fleet learning, with an investment in data quality , rather than quantity. … Read more · 8 min read 48 1 Nina Qi · Oct 27, 2020 Partners with First Transit Working together to provide a safe, reliable, and affordable robotaxi service Scaling Robotaxis to Those Who Need It Most We are excited to announce that is teaming up with First Transit to deploy and operate robotaxis in communities like The Villages . First Transit has 64 years of transportation experience and moves more than 300 million passengers every year. By leveraging such expertise, we will together scale a safe, reliable, and affordable robotaxi service for all. A Collaborative Approach to Robotaxis Building a successful door-to-door robotaxi service requires a lot more than just the underlying technology (although that is certainly the first step). It also requires a vast network of other services, such as on-the-ground operations, fleet management, charging, and insurance. … Read more · 3 min read 17 Oliver Cameron · Aug 26, 2020 Introducing the G3 Robotaxi A better, safer shared vehicle for a COVID-19 world The G3’s ambulance-grade U-VC lighting activated A Better, Safer Shared Vehicle I’m beyond excited to announce the next step towards ’s commitment to delivering on the promise of self-driving cars: The G3, our next-generation robotaxi. The G3 is half the cost of the G2 , ready to thrive in a COVID-19 world, and most importantly, designed to drive without the need for anyone behind the wheel. Combined, these features make it an extremely compelling vehicle to serve our first customers: senior citizens. Beta versions of the G3 are being tested on the road in San Jose today, with production vehicles and commercial driverless to follow next year. … Read more · 8 min read 172 1 Oliver Cameron · Jul 8, 2020 Introducing Telessist Telessist combines the intelligence of a human driver with our self-driving A.I. to handle edge cases The Telessist Pod: A custom-built, automotive-grade workstation for our remote operators A Comprehensive Remote Operations Solution The approach to delivering on the promise of self-driving cars has always been pragmatic and customer-focused. How can we provide value today — not tomorrow or next year — using the state-of-the-art in self-driving technology? We want to ship an autonomous ride-hailing service to the people who need self-driving cars the most now . To accomplish this, we’ve built a suite of self-driving technologies. Commander is our self-driving A.I. designed to drive point-to-point at 25MPH. Shield is our supercharged automatic emergency braking system, complementing Commander and adding extra layers of safety. … Read more · 8 min read 60 Oliver Cameron · May 11, 2020 Partners with FCA to Deliver Fully Driverless Cars Integrating our self-driving technology into a purpose-built Chrysler Pacifica Hybrid An AV-Ready Chrysler Pacifica A Necessary Partnership for Driverless has teamed up with FCA to accelerate ’s deployment of fully driverless cars. We have integrated our self-driving technology into a purpose-built Chrysler Pacifica Hybrid, developed specifically by FCA for the integration of automated technology. Building a self-driving car requires a deep connection between software and hardware, and both pieces need to be considered in tandem. This partnership connects ’s self-driving technology with FCA’s deep expertise in building vehicles to launch truly driverless transportation services. As part of this collaboration, and FCA will jointly adapt and validate the connections between the self-driving software, sensors, and embedded systems. Read more · 3 min read 241 1 Drew Gray · Feb 7, 2020 Active Learning and Why All Data Is Not Created Equal How we can build better machine learning models by using less — but more carefully curated — data Figure 1: A typical fully labeled scene from a self-driving car. The 3D bounding boxes with class type are the labels that make up the training set. This article was written in collaboration with Jennifer Prendki , Founder and CEO of Alectio . The Big Data Labeling Crisis The field of computer vision reached a tipping point when the size and quality of available datasets finally met the needs of theoretical machine learning algorithms. The release of ImageNet , a fully-labeled dataset of 14-million-images, played a critical role. When ImageNet was realized, it was close to impossible for most companies to generate such a large, clean, and (most importantly) labeled dataset for computer vision. … Read more · 12 min read 443 1 Oliver Cameron · Dec 13, 2019 Introducing Shield A supercharged automatic emergency braking system Shield Meet Shield Shield is a supercharged automatic emergency braking system, designed from the ground up for driverless cars. Shield utilizes high-resolution depth sensors, state-of-the-art algorithms, and automotive-grade compute to make intelligent decisions about emergency braking events. We are excited to begin sharing more information about this safety-critical system. Read more · 7 min read 94 1 Drew Gray · Nov 13, 2019 Introducing Deepdrive Unlocking the potential of deep reinforcement learning A snapshot from Deepdrive As ’s CTO, I spend a lot of time...