Automotive autopilot system

Veretelnikov, A.S.; Dmitrienko, N.A. · 2022 · Crossref

DOI: 10.18411/trnio-12-2022-458

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the development and operational concepts of automotive autopilot systems, specifically focusing on unmanned vehicles. The authors highlight that while autopilot technology is established in aviation and shipping due to the relative freedom of those environments, road transport presents unique challenges due to the high density of moving and stationary obstacles requiring real-time trajectory analysis. The study aims to outline the general algorithm for autonomous driving, the role of artificial intelligence and neural networks, the hardware involved, and the current state of automation levels in the industry. The paper describes the core algorithm of an unmanned vehicle as a continuous real-time process divided into four subtasks: localization, perception, prediction, and planning. Localization determines the vehicle’s position using sensor data and high-precision maps containing infrastructure details. Perception involves distinguishing environmental objects to determine their speed and direction. Prediction utilizes trained neural networks to anticipate the movements of other road users. Finally, planning determines the vehicle’s actions and commands control systems based on the previous three stages. The authors note that deep learning enables the identification of patterns in traffic lights, curbs, and other objects. Hardware systems, exemplified by Waymo, combine cameras, sensors, and lidar to analyze the traffic situation. The authors review the six levels of automation defined by the US National Highway Traffic Safety Administration (NHTSA), ranging from Level 0 (no automation) to Level 5 (full autonomy). Level 1 involves advanced driver assistance systems (ADAS) for warnings, while Level 2 allows simultaneous steering and acceleration control, with Tesla cited as a prominent example. Level 3 permits the system to handle basic functions but requires driver intervention for complex maneuvers. Level 4 handles complex situations without requiring driver concentration, and Level 5 entails full AI control, potentially eliminating the steering wheel. The paper states that current development has reached Level 4, with companies like Google (Waymo) and General Motors (Cadillac Super Cruise) leading competitors such as Tesla, Uber, Lyft, and Yandex in test formats. The significance of this research lies in the potential safety improvements offered by autonomous vehicles. The authors cite NHTSA estimates indicating that 94% of serious accidents result from human error or traffic rule violations, such as driving under the influence. By excluding these human factors, autopilot systems can significantly increase road safety. The paper concludes that despite current limitations, major manufacturers are actively investing in this technology, viewing unmanned vehicles as the future of motor transport.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• driverless ads
• situational awareness
• automation surprise

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Theoretical Contribution: conceptual framework