A dynamic fuzzy video compression control algorithm for wireless Advanced Driver Assistance Systems

Created by W.Langdon from gp-bibliography.bib Revision:1.8414

@Article{Prauzek:2025:engappai,

• author = "M. Prauzek and P. Kroemer and J. Konecny and M. Stankus and P. Musilek",
• title = "A dynamic fuzzy video compression control algorithm for wireless Advanced Driver Assistance Systems",
• journal = "Engineering Applications of Artificial Intelligence",
• year = "2025",
• volume = "153",
• pages = "110815",
• keywords = "genetic algorithms, genetic programming, Adaptive compression control, Advanced Driver Assistance Systems, Differential Evolution, Evolutionary Fuzzy Rules, Particle Swarm Optimization, Video compression",
• ISSN = "0952-1976",
• URL = "https://www.sciencedirect.com/science/article/pii/S0952197625008152",
• DOI = "doi:10.1016/j.engappai.2025.110815",
• abstract = "As video-based Advanced Driver Assistance Systems (ADAS) become integral to modern vehicle safety, the demand for reliable, high-performance wireless solutions for retrofitting vehicles have grown. This study introduces a wireless ADAS that employs a novel dynamic video compression control algorithm, integrating a hardware-based Motion Joint Photographic Experts Group (MJPEG) compression engine with adaptive fuzzy logic control strategies. The system dynamically adjusts video compression levels based on real-time conditions such as wireless data rates and available bandwidth, addressing key challenges in maintaining video quality and minimizing latency in wireless environments. The adaptive control is governed by two distinct fuzzy control strategies: Fuzzy Rule-Based (FRB) and Evolutionary Fuzzy Rules (EFR). Both strategies were optimised using nature-inspired algorithms, including Differential Evolution (DE), Particle Swarm Optimisation (PSO), and Genetic Programming (GP). Among these, the EFR-based control was found to offer the best overall performance. Key performance indicators such as compression efficiency, latency, and throughput rates were thoroughly evaluated. Experimental results demonstrated that the EFR-based system provided up to a 35percent improvement in compression efficiency compared to traditional methods, reduced video latency by approximately 20percent, and optimised data throughput. Furthermore, the EFR-based control showcased enhanced generalisation capabilities, outperforming FRB-based control under previously unobserved conditions, which is critical for real-world vehicular applications where network conditions may vary significantly. The implementation of artificial intelligence in the form of EFR significantly enhanced the system's ability to adapt to varying data rates and environmental conditions, making it a promising solution for real-time video compression in computationally constrained embedded systems",

}

Genetic Programming entries for Michal Prauzek Pavel Kromer Jiri Konecny M Stankus Petr Musilek

Citations