Журнал «Современная Наука»

Russian (CIS)English (United Kingdom)
 MOSCOW +7(495)-142-86-81

DEVELOPMENT OF A HYBRID MAPPING ARCHITECTURE FOR GROUND VEHICLE AUTOPILOT SYSTEMS

Yaroschuk Pavel Olegovich  (Postgraduate Student Voronezh State University )

This study focuses on the development and evaluation of a hybrid architecture for map construction in ground vehicle autopilot systems. By combining deep learning, mathematical optimization, and advanced localization algorithms, we synthesized an algorithm designed to estimate the 3D position of dynamic objects, map static environments, and simultaneously localize the ego-object (SLAM). The proposed method demonstrates improved computational speed compared to existing approaches while maintaining competitive accuracy. Unlike methods such as SMOKE, it generates richer scene-understanding data, enabling broader application without additional modifications. The algorithm is highly configurable and human-interpretable, allowing adjustable trade-offs between processing speed and mapping precision. It can be integrated as a component of autopilot systems ranging from simple robotic platforms (e.g., warehouse forklifts) to complex consumer vehicle autonomy (e.g., adaptive cruise control). The solution is self-contained, requiring only fine-tuning of a 2D detector to adapt to specific operational conditions.

Keywords:mathematical optimization, deep learning, slam method, dynamic programming, pnp problem
 

Read the full article …

Citation link:
Yaroschuk P. O. DEVELOPMENT OF A HYBRID MAPPING ARCHITECTURE FOR GROUND VEHICLE AUTOPILOT SYSTEMS // Современная наука: актуальные проблемы теории и практики. Серия: Естественные и Технические Науки. -2025. -№06/2. -С. 254-258 DOI 10.37882/2223-2966.2025.06-2.50
LEGAL INFORMATION:
Reproduction of materials is permitted only for non-commercial purposes with reference to the original publication. Protected by the laws of the Russian Federation. Any violations of the law are prosecuted.
© ООО "Научные технологии"

Terms Publications

Issues of the journals (Archive)

ECONOMICS AND LAW
HUMANITIES
NATURAL AND TECHNICAL SCIENCES
COGNITION

Серия - Natural and Technical Sciences