Applied Intuition’s Post

Pioneering off-road autonomy with AI and ML! 🚜 Our off-road autonomy stack, featured by VentureBeat, showcases our approach to overcoming off-road challenges. Check it out! :)

Pioneering off-road autonomy with AI and ML! 🚜 Our off-road autonomy stack, featured by VentureBeat, showcases our approach to overcoming off-road challenges. Check it out! :)

Pioneering off-road autonomy with AI and ML! 🚜 Our off-road autonomy stack, featured by VentureBeat, showcases our approach to overcoming off-road challenges. Check it out! :)

Pioneering off-road autonomy with AI and ML! 🚜 Our off-road autonomy stack, featured by VentureBeat, showcases our approach to overcoming off-road challenges. Check it out! :)

The curse of rarity – the rarity of safety-critical events in high-dimensional variable spaces – presents significant challenges in ensuring the #safety of #autonomous #vehicles using #deeplearning. Looking at it from distinct perspectives, in a new #comment #naturecommunications Nature Portfolio the authors identify three potential approaches for addressing the issue. Now out 👉 https://lnkd.in/em7Pe5pd.

Why do #autonomousvehicles have trouble navigating edge cases? Some analysts and engineers say that #AI's inability to make cause-effect connections is at the root of the #autonomy industry's challenges to scale the technology safely. Others say all AI needs is more training data. Read about how #causality fits into the quest for safer AVs in Automotive News. https://lnkd.in/eVTCNnct Insights from Alex Oyler, Tim Wong, Srikanth Saripalli, Philip Koopman, Elad Hofstetter

This article from Molly at Automotive News does a great job breaking down one of the toughest challenges to solve in autonomous driving - building #AI systems that can reason and understand causality. It's one of the fundamental limitations of current #machinelearning techniques that hampers the functionality and scalability of #AVs, particularly in mixed traffic (humans+AVs). It could work great in a more controlled environment, but the chaotic nature of traffic today makes for an very challenging operational design domain. That said, there are some incredible opportunities to not only improve safety outcomes through #ADAS technology but also expand productivity in mobility through highway autonomy with the technology we have today. I - and SBD Automotive - are keenly following the development of new machine learning techniques that can help solve this problem, including "causal AI" research and #neuromorphicomputing. Fun fact: SBD Automotive published a report nearly 8 years ago predicting exactly the scalability problem AVs are facing today. If you're interested in a little time capsule (and I know you!), shoot over a DM and I'll share a snippet 😉

Ghost Autonomy's Bold Bet: Harnessing Language Models for Safer Self-Driving-Kyle Wiggers Ghost Autonomy, a startup focused on autonomous driving software, aims to enhance self-driving technology using multimodal large language models (LLMs) with the support of OpenAI and a $5 million investment. Ghost plans to leverage LLMs, which understand both text and images, for complex scene interpretation in self-driving, addressing challenges like navigating construction zones. However, experts express skepticism, viewing the use of LLMs in self-driving as potentially inefficient and premature. Despite criticisms, Ghost's CEO, John Hayes, believes LLMs can improve autonomy by providing reasoning capabilities for diverse driving scenarios, emphasizing ongoing testing and collaboration with automakers to validate and integrate new models. John Hayes #GhostAutonomy #AutonomousDriving #LLMs #OpenAI #SelfDrivingTechnology #AIinAutomotive #SafetyInAutomation #TechInnovation #LanguageModels #FutureOfDriving

Is the dream of driverless cars dead? It's true, there is still a long way to go. As this article quite rightly points out, there have been too many incidents and the progress has been slow. Many companies are pulling back and CEOs are quitting altogether. However, I think the potential is still there and things aren't quite as bleak as the author makes out. What do you think? https://bit.ly/3tfYrFS #AI #Autonomous #AutonomousVehicles #Tesla