- Type
- AI application in transportation
- Key technologies
- Cameras, LiDAR, radar, AI perception and planning
- Standard
- SAE J3016 levels 0–5
- Pioneers
- Waymo, Tesla, Baidu, Mercedes-Benz
- Related
- Computer vision, deep learning, edge AI
- Type
- AI application in transportation
- Key technologies
- Cameras, LiDAR, radar, AI perception and planning
- Standard
- SAE J3016 levels 0–5
- Pioneers
- Waymo, Tesla, Baidu, Mercedes-Benz
- Related
- Computer vision, deep learning, edge AI
An autonomous vehicle (AV) is a vehicle that uses technology to partially or entirely replace the human driver in navigating from an origin to a destination, sensing road conditions, hazards, and traffic and responding accordingly [1]. The Society of Automotive Engineers (SAE) defines six levels of driving automation — from Level 0 (no automation) to Level 5 (full automation with no human oversight) — which are widely used as the framework for regulation and public communication [1][2][3].
History
Research into self-driving vehicles dates to the 1980s, with landmark progress in the DARPA Grand Challenges of 2004–2007, which demonstrated that autonomous ground vehicles could navigate challenging courses. In 2009 Google began its self-driving car project, which was spun off as Waymo in 2016; Waymo launched a commercial robotaxi service, Waymo One, in Phoenix in 2018 and began operating fully driverless ride-hailing in several US cities [6]. Tesla has deployed its Autopilot and Full Self-Driving systems as advanced driver assistance, and by 2025 its vehicles had driven over 10.97 billion miles at Level 2 [6].
In parallel, manufacturers pushed into conditional automation: Mercedes-Benz received regulatory approval for its Drive Pilot Level 3 system in Germany in 2021 and later in the United States, and BMW became the first manufacturer approved for a combination of Level 2 and Level 3 in 2026 [5][6]. In June 2026, the UN Economic Commission for Europe's World Forum for Harmonization of Vehicle Regulations adopted the first global regulations for fully autonomous driving systems (ADS), establishing common safety requirements, validation methods, and data-recording obligations for driverless vehicles [7]. Industry analyses project that adoption will be gradual, with commercial fleets and robotaxis ahead of private ownership [1][4].
Key Concepts
Autonomous vehicles perceive the world through a sensor suite that typically combines cameras, LiDAR, radar, ultrasonic sonar, GPS, and inertial navigation [2][3]. Deep-learning models perform object detection, lane and sign recognition, and scene understanding; planning algorithms then compute trajectories, and control systems execute steering, acceleration, and braking [2][3]. High-definition maps and, increasingly, vehicle-to-everything (V2X) connectivity support navigation and coordination [3].
The SAE levels of driving automation define the division of responsibility between system and driver. Levels 0–2 are driver-assistance systems (e.g., adaptive cruise control and lane keeping) that require an attentive human driver; Level 3 allows the system to drive under specific conditions with the driver ready to take over; Level 4 operates fully autonomously within a defined operational design domain; and Level 5 requires no human input at all [1][2][5].
Applications
Autonomous technology is being deployed in three main use cases: personal vehicles, robotaxis and roboshuttles, and autonomous trucks [4]. Robotaxi operators such as Waymo (United States) and Baidu's Apollo Go (China) run commercial Level 4 services in designated cities, with Waymo reporting over 170 million driverless miles by 2026 [4][6]. Driver-assistance features (Level 2) are now common in production cars, and Level 3 systems are available in limited conditions in Germany, the United States, and elsewhere [5][6]. Beyond passenger transport, autonomous trucks and delivery robots are in commercial pilots, and the technology is expected to reduce accidents, improve mobility for non-drivers, and reshape logistics [1][4].
>See Also
🇲🇾 Malaysia is preparing for autonomous vehicles at the policy level. In August 2026, Digital Minister Gobind Singh Deo said the country's road network was "almost ready" to support AV activity, with network upgrades targeted for completion by the end of 2026, alongside legal changes to permit vehicles to operate without human drivers — existing laws currently require a qualified driver on public roads [8]. The government is also reviewing related areas such as motor insurance for driverless operation [8].
Industry bodies are pushing for capability targets: the Malaysia Automotive, Robotics and IoT Institute (MARii) has reported that Malaysia aims to achieve Level 3 autonomous driving by 2030 [9]. Academics from Monash University Malaysia and Taylor's University have cautioned that systems trained abroad must be adapted to Malaysian conditions — heavy tropical rain can "blind" cameras and disrupt LiDAR — and have urged comprehensive AV regulations, safety-testing frameworks, and independent audits tailored to local road and traffic conditions [10]. These efforts connect to the National AI Office's broader AI agenda and to the growth of Malaysia's automotive and smart-city sectors, with data governance under the PDPA relevant to the collection of driving and location data [8][10].
References
- ↑[Forbes — The Technology Behind Self-Driving Cars, Explained](https://www.forbes.com/sites/technology/article/self-driving-cars)
- ↑[Rambus — SAE levels of automation in cars simply explained](https://www.rambus.com/blogs/driving-automation-levels)
- ↑[Numalis — AI in Self-Driving Cars](https://numalis.com/ai-in-self-driving-cars)
- ↑[World Economic Forum — When will autonomous vehicles and self-driving cars hit the road?](https://www.weforum.org/stories/energy-transition/autonomous-vehicles-technology-future)
- ↑[HERE Technologies — What are the six levels of autonomous driving?](https://www.here.com/learn/blog/6-driving-automation-levels)
- ↑[University of Michigan Center for Sustainable Systems — Autonomous Vehicles Factsheet](https://css.umich.edu/publications/factsheets/mobility/autonomous-vehicles-factsheet)
- ↑[UN News — New global rules clear the road for driverless vehicles (June 2026)](https://news.un.org/en/story/2026/06/1167797)
- ↑[The Star — Malaysia prepares road networks and laws for autonomous vehicles by year-end, says Gobind (13 August 2026)](https://www.thestar.com.my/news/nation/2026/08/13/malaysia-prepares-road-networks-and-laws-for-autonomous-vehicles-by-year-end-says-gobind)
- ↑[MARii — Deputy MITI minister: Malaysia targeting Level 3 autonomous driving by 2030](https://www.facebook.com/MARiiMalaysia/posts/1475550031269152)
- ↑[The Star — The future of transport: Assessing Malaysia's readiness for autonomous cars (November 2025)](https://www.thestar.com.my/tech/tech-news/2025/11/10/the-future-of-transport-assessing-malaysia039s-readiness-for-autonomous-cars)