search
HomeTechnology peripheralsAIL3 will be launched in the first half of next year at the latest: ideal end-to-end autonomous driving and greatly improved performance

L3 will be launched in the first half of next year at the latest: ideal end-to-end autonomous driving and greatly improved performance

Aug 07, 2024 am 04:35 AM
industryworld modelIntelligent drivingideal carMultimodal large modelVLM

Recently, with the rise of generative AI technology, many new car-making forces are exploring new methods of visual language models and world models. End-to-end intelligent driving new technologies seem to have become a common research direction. Last month, Li Auto released the third-generation autonomous driving technology architecture of end-to-end + VLM visual language model + world model. This architecture has been pushed to thousands of people for internal testing. It personifies intelligent driving behavior, improves the information processing efficiency of AI, and enhances the ability to understand and respond to complex road conditions. Li Xiang once said in a public sharing that in the face of rare driving environments that are difficult for most algorithms to identify and process, VLM (Visual Language Model) can systematically improve the capabilities of autonomous driving. This method can be achieved theoretically A breakthrough.

L3 will be launched in the first half of next year at the latest: ideal end-to-end autonomous driving and greatly improved performance

The new generation of autonomous driving systems has greatly increased the upper limit of capabilities - allowing AI to deal with many situations that were difficult to solve in the past, and also lowered the threshold - reducing the need for the size of technology R&D teams, and is expected to allow more people to drive in the near future Get a vastly improved experience in the future.
This set of autonomous driving technology architecture is inspired by the fast and slow system theory of Nobel Prize winner Daniel Kahneman. Simulating human thinking and decision-making processes in the field of autonomous driving also requires "fast systems" and "slow systems" Collaborate. Among them:
・ The fast system (System 1) is good at handling simple tasks and is human intuition formed based on experience and habits; in autonomous driving, it is composed of an end-to-end large model, including perception and planning, which is enough to handle 95% of the problems when driving a vehicle. Routine scenario.
・ The slow system (System 2) is the logical reasoning, complex analysis and computing capabilities formed by humans through deeper understanding and learning; in the autonomous driving system, it is mainly the VLM model, which is used to solve complex or even unknown problems when driving a vehicle Traffic scenes account for about 5% of daily driving scenes.
Last week, at an event held at Li Auto’s Beijing R&D headquarters, Li Auto’s Vice President of Intelligent Driving Lang Xianpeng emphasized that Li Auto’s intelligent driving has now fully integrated into the end-to-end + large model solution, which allows vehicles to understand complex road conditions and traffic rule.
"Both end-to-end and traditional perception decision-making models require a large amount of data for training. One potential problem is that the system will not work well if it encounters unseen scenes," Lang Xianpeng said. "We are exploring the ability of vehicles to think and make decisions like humans."

L3 will be launched in the first half of next year at the latest: ideal end-to-end autonomous driving and greatly improved performance

Li Auto Beijing Headquarters.

Since the second half of last year, Ideal began to adjust its strategy and change its trajectory. In February this year, in the DriveVLM paper submitted by Tsinghua University's Cross-Information Research Institute and Li Auto, researchers applied the visual language model (VLM) that has recently emerged in the field of generative AI and demonstrated extraordinary capabilities in visual understanding and reasoning.

In the industry, this is the first work to propose an autonomous driving speed system. Its method fully combines the mainstream autonomous driving pipeline and a large model pipeline with logical thinking, and is the first to complete the large model work of end test deployment ( Based on NVIDIA Orin platform).

L3 will be launched in the first half of next year at the latest: ideal end-to-end autonomous driving and greatly improved performance

DriveVLM system

DriveVLM consists of a Chain-of-Though (CoT) process with three key modules:

  1. Scenario Description: Use language to describe the driving environment and identify key objects.
  2. Scene Analysis: Dive into the characteristics of key objects and their impact on the ego vehicle.
  3. Hierarchical planning: Step-by-step plan development from meta-action and decision descriptions to waypoints.

These modules correspond to the perception, prediction and planning components in the traditional autonomous driving system process. The difference lies in their ability to handle object perception, intention-level prediction and task-level planning, which have been extremely challenging in the past.

Technical verification

Ideal verification technology is effective in long-tail scenarios:

  • Disassemble real environment data
  • Use generative models to supplement new perspectives
  • Customize changes to weather, time, traffic flow and other conditions

Practical application

Li Auto’s end-to-end model and VLM model run in real time:

  • End-to-end model: higher frame rate
  • VLM model: larger number of parameters, lower frame rate

In complex cities In the scenario, VLM plays a role in situations where decision-making is impossible and delivers decision results and trajectories to the end-to-end model.

End-to-end approach

The end-to-end approach has become a technological watershed, marking the beginning of the real use of AI.

The new generation AI model

The new generation AI model can serve as the question maker:

  • Select the data of users who meet the standard of private car drivers as "real questions"
  • Combined with the world model to generate "simulation questions"

Computing power challenge

車両側での VLM などのモデルの展開は、次のようなコンピューティング能力の課題に直面しています:

  • パラメータの数を最適に保つ
  • 意思決定の待ち時間を改善するためのエンジニアリングの最適化

競争の見通し

Tesla FSD は間もなく国内スマートドライビング分野への参入 新たな競争ステージへの参入:

  • 理想のクルマの目標: エンドツーエンド+VLM自動運転の量産化

The above is the detailed content of L3 will be launched in the first half of next year at the latest: ideal end-to-end autonomous driving and greatly improved performance. For more information, please follow other related articles on the PHP Chinese website!

Statement
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
The AI Skills Gap Is Slowing Down Supply ChainsThe AI Skills Gap Is Slowing Down Supply ChainsApr 26, 2025 am 11:13 AM

The term "AI-ready workforce" is frequently used, but what does it truly mean in the supply chain industry? According to Abe Eshkenazi, CEO of the Association for Supply Chain Management (ASCM), it signifies professionals capable of critic

How One Company Is Quietly Working To Transform AI ForeverHow One Company Is Quietly Working To Transform AI ForeverApr 26, 2025 am 11:12 AM

The decentralized AI revolution is quietly gaining momentum. This Friday in Austin, Texas, the Bittensor Endgame Summit marks a pivotal moment, transitioning decentralized AI (DeAI) from theory to practical application. Unlike the glitzy commercial

Nvidia Releases NeMo Microservices To Streamline AI Agent DevelopmentNvidia Releases NeMo Microservices To Streamline AI Agent DevelopmentApr 26, 2025 am 11:11 AM

Enterprise AI faces data integration challenges The application of enterprise AI faces a major challenge: building systems that can maintain accuracy and practicality by continuously learning business data. NeMo microservices solve this problem by creating what Nvidia describes as "data flywheel", allowing AI systems to remain relevant through continuous exposure to enterprise information and user interaction. This newly launched toolkit contains five key microservices: NeMo Customizer handles fine-tuning of large language models with higher training throughput. NeMo Evaluator provides simplified evaluation of AI models for custom benchmarks. NeMo Guardrails implements security controls to maintain compliance and appropriateness

AI Paints A New Picture For The Future Of Art And DesignAI Paints A New Picture For The Future Of Art And DesignApr 26, 2025 am 11:10 AM

AI: The Future of Art and Design Artificial intelligence (AI) is changing the field of art and design in unprecedented ways, and its impact is no longer limited to amateurs, but more profoundly affecting professionals. Artwork and design schemes generated by AI are rapidly replacing traditional material images and designers in many transactional design activities such as advertising, social media image generation and web design. However, professional artists and designers also find the practical value of AI. They use AI as an auxiliary tool to explore new aesthetic possibilities, blend different styles, and create novel visual effects. AI helps artists and designers automate repetitive tasks, propose different design elements and provide creative input. AI supports style transfer, which is to apply a style of image

How Zoom Is Revolutionizing Work With Agentic AI: From Meetings To MilestonesHow Zoom Is Revolutionizing Work With Agentic AI: From Meetings To MilestonesApr 26, 2025 am 11:09 AM

Zoom, initially known for its video conferencing platform, is leading a workplace revolution with its innovative use of agentic AI. A recent conversation with Zoom's CTO, XD Huang, revealed the company's ambitious vision. Defining Agentic AI Huang d

The Existential Threat To UniversitiesThe Existential Threat To UniversitiesApr 26, 2025 am 11:08 AM

Will AI revolutionize education? This question is prompting serious reflection among educators and stakeholders. The integration of AI into education presents both opportunities and challenges. As Matthew Lynch of The Tech Edvocate notes, universit

The Prototype: American Scientists Are Looking For Jobs AbroadThe Prototype: American Scientists Are Looking For Jobs AbroadApr 26, 2025 am 11:07 AM

The development of scientific research and technology in the United States may face challenges, perhaps due to budget cuts. According to Nature, the number of American scientists applying for overseas jobs increased by 32% from January to March 2025 compared with the same period in 2024. A previous poll showed that 75% of the researchers surveyed were considering searching for jobs in Europe and Canada. Hundreds of NIH and NSF grants have been terminated in the past few months, with NIH’s new grants down by about $2.3 billion this year, a drop of nearly one-third. The leaked budget proposal shows that the Trump administration is considering sharply cutting budgets for scientific institutions, with a possible reduction of up to 50%. The turmoil in the field of basic research has also affected one of the major advantages of the United States: attracting overseas talents. 35

All About Open AI's Latest GPT 4.1 Family - Analytics VidhyaAll About Open AI's Latest GPT 4.1 Family - Analytics VidhyaApr 26, 2025 am 10:19 AM

OpenAI unveils the powerful GPT-4.1 series: a family of three advanced language models designed for real-world applications. This significant leap forward offers faster response times, enhanced comprehension, and drastically reduced costs compared t

See all articles

Hot AI Tools

Undresser.AI Undress

Undresser.AI Undress

AI-powered app for creating realistic nude photos

AI Clothes Remover

AI Clothes Remover

Online AI tool for removing clothes from photos.

Undress AI Tool

Undress AI Tool

Undress images for free

Clothoff.io

Clothoff.io

AI clothes remover

Video Face Swap

Video Face Swap

Swap faces in any video effortlessly with our completely free AI face swap tool!

Hot Tools

WebStorm Mac version

WebStorm Mac version

Useful JavaScript development tools

mPDF

mPDF

mPDF is a PHP library that can generate PDF files from UTF-8 encoded HTML. The original author, Ian Back, wrote mPDF to output PDF files "on the fly" from his website and handle different languages. It is slower than original scripts like HTML2FPDF and produces larger files when using Unicode fonts, but supports CSS styles etc. and has a lot of enhancements. Supports almost all languages, including RTL (Arabic and Hebrew) and CJK (Chinese, Japanese and Korean). Supports nested block-level elements (such as P, DIV),

EditPlus Chinese cracked version

EditPlus Chinese cracked version

Small size, syntax highlighting, does not support code prompt function

DVWA

DVWA

Damn Vulnerable Web App (DVWA) is a PHP/MySQL web application that is very vulnerable. Its main goals are to be an aid for security professionals to test their skills and tools in a legal environment, to help web developers better understand the process of securing web applications, and to help teachers/students teach/learn in a classroom environment Web application security. The goal of DVWA is to practice some of the most common web vulnerabilities through a simple and straightforward interface, with varying degrees of difficulty. Please note that this software

SublimeText3 English version

SublimeText3 English version

Recommended: Win version, supports code prompts!