OZO. Y. Zelinskyi

O. Y. Zelinskyi

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


O. Y. Zelinskyi has a total of 3 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    ADAPTIVE MULTI-TARGET RECONNAISSANCE BY UAV SWARMS IN EW ENVIRONMENTS: A HYBRID AUCTION-PSO APPROACH

    by O. Y. Zelinskyi, I. O. Rabiichuk, P. V. Serdyuk, A. V. Fechan

    The rapid proliferation of Unmanned Aerial Vehicle (UAV) swarms in modern combat theaters has necessitated the development of robust autonomous control systems capable of operating effectively in environments heavily saturated with Electronic Warfare (EW) assets. Traditional navigation paradigms frequently fail under such conditions due to global navigation satellite system spoofing and the systematic disruption of remote operator control channels. This paper proposes a novel hybrid control architecture for multi-target reconnaissance that synergistically integrates a market-based Auction Algorithm for strategic target allocation and a modified Particle Swarm Optimization (PSO) algorithm for tactical path planning. The integration of these distinct logical levels utilizes a Finite State Machine (FSM) mechanism, which ensures strict deterministic switching between cruising, pursuit, and evasive flight modes during the mission. The strategic allocation phase employs a complex non-linear cost function designed to evaluate multiple parameters simultaneously. This function accounts not only for the geometric distance to the target but also for the current energy states of individual agents and the predicted environmental threat levels calculated through discrete spatial sampling. Furthermore, the tactical navigational logic incorporates advanced bio-inspired mechanisms to enhance survivability. The modified algorithm introduces adaptive inertia weights and tangential repulsive forces, which collectively empower the drones to autonomously execute sophisticated "sliding mode" maneuvers directly along the boundaries of active jamming zones. The research methodology included a rigorous comparative analysis using Monte Carlo simulations within a comprehensive 60×60 kilometer operational area. This simulated environment featured highly rugged terrain and modeled a swarm of 20 UAVs tasked with engaging 100 distant targets while navigating through 15 dynamically active EW suppression zones. The simulation results conclusively demonstrate that the proposed hybrid approach significantly outperforms classical greedy algorithms. Specifically, the hybrid system reduced the integral damage sustained from EW irradiation by 18.4% and decreased the total time spent within high-risk danger zones by 12.2%. Moreover, the analysis revealed a particularly significant "energy paradox" during the autonomous operations. Despite a 3.4% increase in the average flight distance caused by the necessary evasive maneuvers, the swarm achieved superior overall energy efficiency by consuming 10.79% of battery resources compared to the 10.88% consumed by straight-line trajectories. Ultimately, these findings prove the high effectiveness of the proposed hybrid method for executing autonomous reconnaissance missions under conditions of extreme uncertainty. The developed control architecture offers a highly viable and practical pathway for the future deployment of resilient unmanned aerial systems in actively contested airspace.