E661: George Hotz founder Comma.ai on vision & tech behind his Tesla rival, self-driving car startup

TL;DR

Comma.ai is developing a level three autonomous driving system that aims to make driving safer and more convenient, with a target price of under $1,000.

Transcript

this weekend startups is brought to you by Walker corporate law a boutique Law Firm specializing in the representation of entrepreneurs visit them at Walker corporat law.com and Squarespace start building your website today at squarespace.com enter offer code twist at checkout to get 10% off hey everybody welcome to this weekend startups I'm your h... Read More

Key Insights

• 😥 Comma.ai's goal is to make autonomous driving systems affordable and accessible by targeting a price point of under $1,000.
• 🎥 The company believes that cameras and advanced processing technology are crucial for autonomous driving, rather than relying on sensors like LiDAR.
• 🧑‍🏭 The potential for autonomous driving is significant in the trucking industry, as the cost of labor is a major factor and highway driving is more predictable and controlled.
• 👤 The future of autonomous driving may involve a combination of user installable kits, AI-driven simulation, and improved camera technology.

Questions & Answers

Q: When will level four autonomous driving be achieved in major cities in the United States?

Comma.ai does not build level four systems but focuses on level three autonomy. Level four autonomy, where the car can fully operate without human intervention, is a regulatory and technical challenge that may take longer to accomplish.

Q: What are the core components required for Comma.ai's autonomous driving technology?

Currently, Comma.ai's system uses a single camera and a single radar. They plan to support cars with pre-installed radars and require users to install a camera. The camera processes images to determine the car's path, lane position, and the presence of other vehicles.

Q: What are the challenges in building autonomous driving systems for cities and how does it differ from highway driving?

City driving requires more periphery awareness and currently poses a greater challenge than highway driving. Comma.ai is exploring city driving but acknowledges that it is a different problem and may take longer to develop a reliable system for city environments.

Q: Is it possible that self-driving cars can communicate with each other to improve safety and optimize traffic flow?

Comma.ai does not believe that vehicle-to-vehicle communication is a practical solution. They focus on camera and radar inputs for sensing and processing the environment, rather than relying on complex communication systems.

Summary & Key Takeaways

• Comma.ai builds driver assistance systems that aim to achieve level three autonomy, in which the car takes over multiple aspects of driving but the driver must still be present and attentive.

• Their system uses a combination of cameras and radars to process images and predict the car's path, detect lanes, and avoid collisions with other vehicles and pedestrians.

• The goal is to launch a product that can be easily installed in existing vehicles, with a focus on user installability and affordability.