All Segments
Sterling Anderson, Co-Founder, Aurora - MIT Self-Driving Cars
29 segments available
This is a talk by Sterling Anderson for course 6.S094: Deep Learning for Self-Driving Cars (2018 version). Sterling is the Co-Founder of Aurora, previously head of the Tesla Autopilot team. This class is free and open to everyone. It is an introduction to the practice of deep learning through the applied theme of building a self-driving car. INFO: Course website: https://selfdrivingcars.mit.edu Contact: deepcars@mit.edu CONNECT: - If you enjoyed this video, please subscribe to this channel. - AI Podcast: https://lexfridman.com/ai/ - LinkedIn: https://www.linkedin.com/in/lexfridman - Twitter: https://twitter.com/lexfridman - Facebook: https://www.facebook.com/lexfridman - Instagram: https://www.instagram.com/lexfridman - Slack: https://deep-mit-slack.herokuapp.com LINKS: Playlist: https://goo.gl/SLCb1y Lecture 1: Deep Learning - https://youtu.be/-6INDaLcuJY Lecture 2: Self-Driving Cars - https://youtu.be/_OCjqIgxwHw Lecture 3: Deep Reinforcement Learning - https://youtu.be/MQ6pP65o7OM Lecture 4: Computer Vision - https://youtu.be/CLOAswsxudo Lecture 5: Deep Learning for Human Sensing - https://youtu.be/Z2GfE8pLyxc Guest talk: Sacha Arnoud, Waymo - https://youtu.be/LSX3qdy0dFg Guest talk: Emilio Frazolli, nuTonomy - https://youtu.be/dWSbItd0HEA Guest talk: Sterling Anderson, Aurora - https://youtu.be/HKBhP9JISF0 2017: Guest talk: Sertac Karaman, MIT - https://youtu.be/0fLSf3NO0-s Guest talk: Chris Gerdes, Stanford - https://youtu.be/LDprUza7yT4
Segments Timeline
Introducing Sterling Anderson
Sterling Anderson, co-founder of Aurora and former head of Tesla's Autopilot team, introduces himself and shares his journey from MIT to the forefront of self-driving technology. He highlights his academic background and the evolution of the self-driving community over the past decade.
"today we have sterling Anderson he's the co-founder of Aurora an exciting new self-driving car company previously he was the head of the Tesla auto pilot team that brought both the first the second ge..."
A Decade of Change in Self-Driving
Anderson reflects on the significant changes in the self-driving industry over the last ten years. He discusses his experiences and the lessons learned during his journey, setting the stage for a deeper exploration of the future of self-driving technology and his company, Aurora.
"[Applause] thank you it's good to be here I was telling Lex just before I think it's been a little while since I've been back after the Institute and it's great to be here I want to apologize in advan..."
From MIT to Tesla
Sterling shares his transition from MIT to Tesla, where he led the Model X program and later the Autopilot initiative. He discusses the introduction of new safety and convenience features, emphasizing the innovative approaches taken during his time at Tesla.
"I'd chat with you a little bit about my journey over the last decade it's been just over ten years since I was at MIT a lot has changed a lot has changed for the better in the self-driving community a..."
The Intelligent Co-Pilot
Anderson introduces the concept of the Intelligent Co-Pilot, developed to address safety issues in human driving. He explains the methodology behind this innovation, focusing on optimizing vehicle control through a unique approach that prioritizes human-machine interaction.
"10 years as Lex mentioned when I started MIT I worked with Carly on Yemma Amelio Fazoli's John Leonard a few others on some of these sort of shared adaptive automation approaches I'll talk a little bi..."
Innovative Control Strategies
In this segment, Anderson elaborates on the innovative control strategies used in the Intelligent Co-Pilot system. He discusses how the system identifies and optimizes vehicle trajectories while ensuring safety and stability, highlighting the importance of human factors in automated driving.
"a little bit about that so to start out with when I KN OIT was 2007 the DARPA urban challenge is were well underway at that stage and one of the things that we wanted to do is find a way to address so..."
Simulation and Human Interaction
Sterling describes a simulation that demonstrates the effectiveness of the Intelligent Co-Pilot. He contrasts the performance of a human driver with and without the co-pilot, showcasing how the system enhances stability and safety during driving scenarios.
"moment an environment that's pockmarked by objects by their vehicles by pedestrians etc if you were to create the Voronoi diagram through that environment you would have a set of each unique set of pa..."
Understanding Human Feedback
Anderson discusses the challenges of providing natural feedback to human drivers in automated systems. He shares insights from experiments on how different feedback mechanisms can affect driver performance and mental models, emphasizing the need for intuitive interactions.
"dimensional tube in state space those constraints being spatial imagine for a moment edges of the roadway or you know circumventing various objects in the roadway imagine them also being dynamic right..."
Engaging the Co-Pilot
In this segment, Sterling presents findings from tests involving the Intelligent Co-Pilot. He highlights the significant reduction in collision rates when the co-pilot is engaged, demonstrating its effectiveness in enhancing driver safety and performance.
"itself in a state where that forward simulated optimal trajectory is very near the limits of what the vehicle and it can actually handle we will have transition control fully to the to the vehicle to ..."
Challenges in Human-Machine Collaboration
Anderson explores the complexities of human-machine collaboration in driving. He discusses the importance of maintaining a driver's mental model and the various strategies tested to ensure seamless interaction between human drivers and automated systems.
"the blue vehicles case it's the exact same human driver model now with the copilot in the loop you'll notice that as as this scenario continues what you see here on the left is the green is in this gr..."
DARPA and Unmanned Vehicles
Sterling shares his experiences with DARPA's programs focused on unmanned ground vehicles. He discusses the application of the Intelligent Co-Pilot in military contexts and the insights gained from testing human operators in remote vehicle control scenarios.
"backwards torque to what the vehicle is doing so the car starts to turn right human will feel the wheel turn left they'll see it start to turn left is that more confusing or less confusing they're hum..."
Performance Metrics and Outcomes
In the final segment, Anderson summarizes the performance metrics observed during the tests with the Intelligent Co-Pilot. He highlights the improvements in speed and safety, reinforcing the potential of automated systems to enhance driving experiences.
"bounding various cones and what we did is we set up about 20 drivers 2020 test subjects looking at this this troll screen and operating the vehicle through this track and we set this up as a race with..."
Perception of Control in Autonomous Driving
Anderson shares intriguing findings from his research on driver control perception when using a copilot. Despite a 43% reduction in actual control, drivers reported feeling 12% more in control, illustrating the psychological aspects of human-machine interaction in self-driving cars.
"field as fast as possible they couldn't they had no line-of-sight connection the vehicle and we played with some things on their interface we did you know we caused it to drop out occasionally we dela..."
Founding Aurora: A New Era in Self-Driving
Sterling Anderson reflects on the founding of Aurora, a company aimed at advancing self-driving technology. He discusses the convergence of automotive innovation, machine learning, and the need for a non-threatening business model to accelerate the deployment of autonomous vehicles.
"little bit on sort of the human psyche in terms of you know they were reporting the vehicle was doing what I wanted to do maybe not what I told it to do which was which was kind of fun observation and..."
The Future of Self-Driving Technology
In this segment, Anderson outlines the unique opportunities in the self-driving industry, driven by advancements in machine learning and sensing technologies. He emphasizes the importance of collaboration with automakers to develop scalable self-driving platforms.
"logo it may look familiar to you we turned that into we at the time it intended to roll this out across various automakers in their operations at the time very few saw self-driving as a technology was..."
Partnerships with Major Automakers
Anderson discusses Aurora's partnerships with Volkswagen and Ondine Motor Company, highlighting the collaborative efforts to integrate self-driving technology into their vehicles. He emphasizes the significance of these partnerships in scaling technology globally.
"that was something that didn't exist ten years ago two significant advances have been made in you know some of these machine learning techniques in particular deep learning and other neural network ne..."
Impact of Self-Driving Cars on Society
Sterling Anderson addresses the societal implications of self-driving technology, including increased safety, improved access to transportation, and efficient use of roadways. He underscores the potential of autonomous vehicles to transform urban mobility and enhance quality of life.
"important the important elements of building Lexus is Lex before coming out here what this group would be most interested in hearing one of the things that he mentioned was what does it take to build ..."
Lidar vs. Other Technologies in AVs
In response to a question about lidar usage in self-driving cars, Anderson shares Aurora's approach to leveraging multiple modalities for safe and efficient market entry. He emphasizes the importance of a balanced technology strategy in developing autonomous vehicles.
"forward so with that maybe I'll maybe I'll stop with this and we can go to go to questions I am Wayne - hello thanks for coming um I'm a question a lot of self-driving car companies are making extensi..."
The Future of Car Enthusiasts in an AV World
Anderson reflects on the impact of autonomous vehicles on car enthusiasts and personal vehicle ownership. He envisions a shift towards leisure driving as a sport, highlighting the potential for a new relationship between people and their vehicles in an autonomous future.
"talk about Tesla too much in terms of our specific any anything that wasn't public information I'm not going to get into you I will say that for Aurora we believe that the right approach is getting th..."
Commercialization and Partnerships
Anderson clarifies that Aurora is not yet commercializing self-driving technology but is focusing on partnerships and pilot programs. He discusses the importance of collaborating with automotive partners who understand their customers and distribution networks.
"I think there's a place for both yeah yeah yeah congratulations on the partnership that was announced recently I think so I have a two-part question the first one is so we heard last week from I think..."
Economic and Technological Bottlenecks
In this segment, Anderson addresses the economic and technological bottlenecks in self-driving vehicle development. He explains how the economics of operating self-driving vehicles in shared networks are viable, while also highlighting the challenges in forecasting the behavior of other road users.
"technology and I don't want to get too much into you know the nuances of that business model I will say that it is is one that's done in very close partnership with our automotive partners because you..."
Advancements in Sensor Technology
Anderson discusses the potential for reducing costs in sensor technology and the importance of developing high-capability systems for initial deployment. He emphasizes the automotive industry's ability to drive down costs as technology matures.
"companies today but a handful I think two questions one is in invariably a new product development there's typically two types of bottlenecks there's a technological bottleneck and an economic bottlen..."
Challenges in Predicting Behavior
Anderson elaborates on the challenges of predicting the intent and future behaviors of other road users. He highlights the need for advanced machine learning techniques to improve the forecasting capabilities of self-driving systems.
"let me start with the economic side of it and just to get that at the wake is a little bit quicker answer the economics of operating a self-driving vehicle and a shared network today would close that ..."
Security in Autonomous Vehicles
Addressing ethical concerns, Anderson discusses the importance of security in self-driving technology. He emphasizes the need for ongoing development efforts to protect against malicious actions and the necessity of best practices in system architecture.
"automotive community has shown themselves to be exceptional at driving those costs down and so I expect them to come way down to your other question technological bottlenecks and challenges one of the..."
Redefining Vehicle Design
Anderson explores the opportunities self-driving cars present for reimagining vehicle design. He discusses how the integration of self-driving technology can lead to new user experiences and changes in vehicle interiors, ultimately enhancing safety and functionality.
"with again so this question isn't necessarily about the development of self-driving cars but more like an ethics question when you're putting human lives into like the hands of software isn't there al..."
Testing and Safety in Self-Driving Features
In this segment, Anderson explains the statistical evaluation process for launching new self-driving features, such as throttle control. He emphasizes the challenge of capturing every possible scenario and the importance of ensuring that self-driving systems can perform safely at a level better than the average human driver.
"for passive safety systems so yes I have a question about the go no-go tests that you conduct for certain features like you mentioned the throttle control where you slow down the throttle assuming tha..."
Navigating Job Displacement from Automation
Anderson discusses the societal implications of self-driving technology, particularly the potential displacement of jobs in the transportation sector. He stresses the need for proactive discussions and strategies to transition affected workers into new roles as self-driving systems become more prevalent.
"that threshold I think it the opportunity to launch is is real and you should seriously consider it so thank you for your talk today first and my question is self-driving seems to be able to ultimatel..."
Collaborative Partnerships in Self-Driving Tech
Anderson shares insights on Aurora's partnerships with major automakers like Volkswagen and Hyundai. He explains the importance of collaboration in bringing self-driving technology to market quickly and safely, emphasizing a shared vision for innovation across the automotive industry.
"home for it now I I'm kasha from the Sloan School my question was more about your business model again with partnering with both VW and he and a and you're just perspective and how you were able to ef..."
The Future of Vehicle Communication Networks
In this segment, Anderson discusses the potential for vehicle-to-vehicle and vehicle-to-infrastructure communication networks. He highlights the benefits of sharing information among vehicles to enhance safety and performance, while acknowledging the challenges of scaling these systems effectively.
"Aurora was started to do hi I had a question now that I see a lot of companies are coming up with self-driving cars right so most of the cars are pretty much all the technology is bound only to the ca..."
Optimizing Traffic with Coordinated Self-Driving Vehicles
Anderson concludes by discussing the advantages of coordinating self-driving vehicles for improved traffic management and infrastructure utilization. He emphasizes the value of understanding demand patterns and optimizing vehicle allocation to enhance mobility services and reduce congestion.
"solved thank you for your talk my question is what's your opinion about cooperation of self-driving vehicles so maybe I think if you can control a group of self-driving vehicles at the same time you c..."