To address the larger numbers of small objects and the issues of occlusion and clustering in UAV aerial photography, which can lead to false positives and missed detections, we propose an improved small object detection algorithm for UAV aerial scenarios called YOLOv8 with tiny prediction head and Space-to-Depth Convolution (HSP-YOLOv8). Firstly, a tiny prediction head specifically for small targets is added to provide higher-resolution feature mapping, enabling better predictions. Secondly, we designed the Space-to-Depth Convolution (SPD-Conv) module to mitigate the loss of small target feature information and enhance the robustness of feature information. Lastly, soft non-maximum suppression (Soft-NMS) is used in the post-processing stage to improve accuracy by significantly reducing false positives in the detection results. In experiments on the Visdrone2019 dataset, the improved algorithm increased the detection precision mAP0.5 and mAP0.5:0.95 values by 11% and 9.8%, respectively, compared to the baseline model YOLOv8s.
Toward the minimal weighted vertex cover (MWVC) in agent-based networking systems, this paper recasts it as a potential game and proposes a distributed learning algorithm based on relaxed greed and finite memory. With the concept of convention, we prove that our algorithm converges with probability 1 to Nash equilibria, which serve as the bridge connecting the game and the MWVC. More importantly, an additional degree of freedom is also provided for equilibrium refinement, such that increasing memory lengths and mutation probabilities contributes to the improvement of system-level objectives. Comparisons with typical methods, centralized and distributed, demonstrate the advantage of our algorithm for both weighted and unweighted versions. This paper not only provides a useful tool for the MWVC problem in decentralized environments but also paves an effective way for distributed coordination and optimization that could be modeled as potential games.
Toward the global optimality of the SET K-COVER problem in wireless sensor networks, we view each sensor node as a rational player and propose a time variant log-linear learning algorithm (TVLLA) that relies on local information only. By defining the local utility as the normalized area covered by one node alone, we formulate the problem as a spatial potential game. The resulting optimal Nash equilibria correspond to the optimal partition. Such equilibria are obtained by designing a time varying parameter β that approaches infinity with time. Using inhomogeneous Markov chain theory, we prove that the TVLLA guarantees convergence to the optimal solution with probability 1. Comparison results against traditional methods demonstrate that the algorithm can also provide better near-optimal solutions in a reasonable computation time than the state-of-the-art. Our findings pave a new way to reach the global optimality of the SET K-COVER problem in a distributed manner as well as other potential games from the view of self-organized optimization.
2 November 2021 in IEEE Transactions on Systems, Man, and Cybernetics: Systems
Toward better approximation for the minimum-weighted vertex cover (MWVC) problem in multiagent systems, we present a distributed algorithm from the perspective of learning in games. For self-organized coordination and optimization, we see each vertex as a potential game player who makes decisions using local information of its own and the immediate neighbors. The resulting Nash equilibrium is classified into two categories, i.e., the inferior Nash equilibrium (INE) and the dominant Nash equilibrium (DNE). We show that the optimal solution must be a DNE. To achieve better approximation ratios, local rules of perturbation and weighted memory are designed, with the former destroying the stability of an INE and the latter facilitating the refinement of a DNE. By showing the existence of an improvement path from any INE to a DNE, we prove that when the memory length is larger than 1, our algorithm converges in finite time to DNEs, which could not be improved by exchanging the action of a selected node with all its unselected neighbors. Moreover, additional freedom for solution efficiency refinement is provided by increasing the memory length. Finally, intensive comparison experiments demonstrate the superiority of the presented methodology to the state of the art, both in solution efficiency and computation speed.
20 January 2021 in IEEE Transactions on Systems, Man, and Cybernetics: Systems
Area coverage and network lifetime are two contradictory issues to the architecture development of a wireless sensor network (WSN). A satisfactory balance could be achieved by deploying abundant sensor nodes randomly and dividing them into $k$ exclusive cover sets. Toward self-organized partition with higher efficiency, we address the problem from the perspective of networked potential games and propose a memorial mixed-response algorithm (MMRA), which is implemented in a distributed and synchronous manner. Being viewed as a game player, each sensor node first updates its memory using a temporary action, which is generated by following a mixed response rule. After this, the coordination evolves into the next iteration by each player randomly drawing an action from its memory with equal probabilities. We prove that our algorithm converges with probability 1 to a convention of Nash equilibria, with the worst approximation ratio strictly larger than 0.5. Moreover, it is also found that a tradeoff between solution efficiency and computation time could be achieved via the adjustment of the amount of randomness introduced via the memory length $m$ as well as the probability $p_{m}$ , where better partition results are more likely to be generated using a larger $m$ and smaller $p_{m}$ . Comparisons with existing distributed methods demonstrate the superiority of our method in terms of solution refinement as well as convergence speed.
This study primarily studies the shortest-path planning problem for unmanned aerial vehicle (UAV) formations under uncertain target sequences. In order to enhance the efficiency of collaborative search in drone clusters, a compensation look-ahead algorithm based on optimizing the four-point heading angles is proposed. Building upon the receding-horizon algorithm, this method introduces the heading angles of adjacent points to approximately compensate and decouple the triangular equations of the optimal trajectory, and a general formula for calculating the heading angles is proposed. The simulation data indicate that the model using the compensatory look forward algorithm exhibits a maximum improvement of 12.9% compared to other algorithms. Furthermore, to solve the computational complexity and sample size requirements for optimal solutions in the Dubins multiple traveling salesman model, a path-planning model for multiple UAV formations is introduced based on the Euclidean traveling salesman problem (ETSP) pre-allocation. By pre-allocating sub-goals, the model reduces the computational scale of individual samples while maintaining a constant sample size. The simulation results show an 8.4% and 17.5% improvement in sparse regions for the proposed Euclidean Dubins traveling salesman problem (EDTSP) model for takeoff from different points.