Mir Feroskhan’s research while affiliated with Nanyang Normal University and other places

Safe navigation of drones in the presence of adversarial physical attacks from multiple pursuers is a challenging task. This paper proposes a novel approach, asynchronous multistage deep reinforcement learning (AMS-DRL), to train adversarial neural networks that can learn from the actions of multiple evolved pursuers and adapt quickly to their behavior, enabling the drone to avoid attacks and reach its target. Specifically, AMS-DRL evolves adversarial agents in a pursuit-evasion game where the pursuers and the evader are asynchronously trained in a bipartite graph way during multiple stages. Our approach guarantees convergence by ensuring Nash equilibrium among agents from the game-theory analysis. We evaluate our method in extensive simulations and show that it outperforms baselines with higher navigation success rates. We also analyze how parameters such as the relative maximum speed affect navigation performance. Furthermore, we have conducted physical experiments and validated the effectiveness of the trained policies in real-time flights. A success rate heatmap is introduced to elucidate how spatial geometry influences navigation outcomes. Project website: https://github.com/NTU-ICG/AMS-DRL-for-Pursuit-Evasion .

This paper addresses the pursuit control problem for multi-agent systems, aiming to ensure collision-free tracking under input saturation and external disturbances. We propose a novel Control Barrier Function (CBF)-Safe Reinforcement Learning (CSRL) algorithm, which integrates model-free reinforcement learning with a safety filter to guarantee system safety. The framework introduces an input-constrained CBF that dynamically adjusts control bounds, enabling robust target tracking even during evasive maneuvers. A safety filter is designed to transform unsafe RL actions into safe control signals by solving a Quadratic Program (QP), ensuring the safety for sensing, collision avoidance, and input constraints of pursuers. Theoretical analysis proves the feasibility of the CBF-QP using the Karush-Kuhn-Tucker (KKT) conditions. Simulation results validate the effectiveness of the CSRL algorithm, demonstrating its ability to handle complex pursuit scenarios while maintaining safety and improving control performance.

Avian perching maneuvers are one of the most frequent and agile flight scenarios, where highly optimized flight trajectories, produced by rapid wing and tail morphing that generate high angular rates and accelerations, reduce kinetic energy at impact. While the behavioral, anatomical, and aerodynamic factors involved in these maneuvers are well described, the underlying control strategies are poorly understood. Here, we use optimal control methods on an avian-inspired drone with morphing wing and tail to test a recent hypothesis derived from perching maneuver experiments of Harris’ hawks that birds minimize the distance flown at high angles of attack to dissipate kinetic energy before impact. The resulting drone flight trajectories, morphing sequence, and kinetic energy distribution resemble those measured in birds. Furthermore, experimental manipulation of the wings that would be difficult or unethical with animals reveals the morphing factors that are critical for optimal perching maneuver performance of birds and morphing-wing drones.

Obstacle avoidance under constrained visual perception presents a significant challenge, requiring rapid detection and decision-making within partially observable environments, particularly for unmanned aerial vehicles (UAVs) maneuvering agilely in three-dimensional space. Compared to traditional methods, obstacle avoidance algorithms based on deep reinforcement learning (DRL) offer a better comprehension of the uncertain operational environment in an end-to-end manner, reducing computational complexity and enhancing flexibility and scalability. However, the inherent trial-and-error learning mechanism of DRL necessitates numerous iterations for policy convergence, leading to sample inefficiency issues. Meanwhile, existing sample-efficient obstacle avoidance approaches that leverage imitation learning often heavily rely on offline expert demonstrations, which are not always feasible in hazardous environments. To address these challenges, we propose a novel obstacle avoidance approach based on Non-Expert Policy Enhanced DRL (NPE-DRL). This approach integrates a fundamental DRL framework with prior knowledge derived from a non-expert policy-guided imitation learning. During the training phase, the agent starts by online imitating the actions generated by the non-expert policy during interactions and progressively shifts toward autonomously exploring the environment to generate the optimal policy. Both simulation and physical experiments validate that our approach improves sample efficiency and achieves a better exploration-exploitation balance in both virtual and real-world flights. Additionally, our NPE-DRL-based obstacle avoidance approach shows better adaptability in complex environments characterized by larger scales and denser obstacle configurations, demonstrating a significant improvement in UAVs’ obstacle avoidance capability. Code available at https://github.com/zzzzzyh111/NonExpert-Guided-Visual-UAV-Navigation-Gazebo .</p

The rapid advancement of drone technology has significantly impacted various sectors, including search and rescue, environmental surveillance, and industrial inspection. Multidrone systems offer notable advantages such as enhanced efficiency, scalability, and redundancy over single-drone operations. Despite these benefits, ensuring resilient formation control in dynamic and adversarial environments, such as under communication loss or cyberattacks, remains a significant challenge. Classical approaches to resilient formation control, while effective in certain scenarios, often struggle with complex modeling and the curse of dimensionality, particularly as the number of agents increases. This paper proposes a novel, learning-based formation control for enhancing the adaptability and resilience of multidrone formations using graph attention networks (GATs). By leveraging GAT's dynamic capabilities to extract internode relationships based on the attention mechanism, this GAT-based formation controller significantly improves the robustness of drone formations against various threats, such as Denial of Service (DoS) attacks. Our approach not only improves formation performance in normal conditions but also ensures the resilience of multidrone systems in variable and adversarial environments. Extensive simulation results demonstrate the superior performance of our method over baseline formation controllers. Furthermore, the physical experiments validate the effectiveness of the trained control policy in real-world flights.