target-relative guidance

This paper addresses the problem of synchronized three-dimensional interception of stationary targets by leaderless multi-interceptor systems, confronting challenges including lateral-acceleration-only actuation, strong coupling between pitch and yaw channels, dynamically switching communication topologies, and time-varying uncertainty in time-to-go (TTG) estimation. To tackle these, we propose a coupled three-dimensional cooperative guidance law: (i) a distributed consensus protocol over switching dynamic graphs ensures state coordination; (ii) instantaneous three-dimensional lateral acceleration commands are directly optimized—bypassing conventional decoupling and thus preserving guidance performance; and (iii) a time-varying TTG estimation error compensation term is incorporated to guarantee prescribed-time convergence under affine constraints. Simulation results demonstrate high-precision simultaneous impact across diverse initial configurations, confirming the method’s robustness against uncertainties and its capability for real-time cooperative engagement.

This paper addresses the challenge of autonomous interception of non-cooperative, dynamic targets by heterogeneous mobile platforms—unmanned aerial vehicles, ground vehicles, and spacecraft—under conditions of no global localization, limited field-of-view, and frequent occlusions. Methodologically, it proposes a monocular vision–driven general interception framework integrating extended Kalman filter–based relative pose estimation, history-conditioned target trajectory prediction, and real-time constrained convex optimization for receding-horizon motion planning. It is the first work to systematically validate cross-platform interception feasibility under weak observability and to establish a unified kinematic adaptation paradigm. Experiments demonstrate sub-0.15 m interception error, >94% success rate, and real-time execution on embedded platforms such as Jetson Orin. The core contribution is the first general-purpose visual interception architecture designed for multiple robot classes, achieving simultaneous robustness against sensing limitations and computational efficiency.

This work addresses the problem of local safe navigation for Ackermann-steered robots in mapless environments without global goals. The authors propose a real-time, perception-only obstacle avoidance method that identifies the largest open sector ahead to determine a safe heading and constructs left–right boundary constraints. A convex quadratic program is then employed to maximize clearance between the vehicle and surrounding obstacles, and a feedback linearization controller tracks the resulting smooth reference trajectory. The approach achieves high computational efficiency while ensuring both safety and trajectory smoothness. Experimental results demonstrate that, compared to existing exploration-based planners, the proposed method significantly reduces computation time and yields safer, more reliable paths.

This work addresses the challenge of efficiently searching for and capturing multiple persistently drifting targets—such as debris or distressed objects—in dynamic aquatic environments. The authors propose a Model Predictive Path Integral (MPPI) planning framework that integrates spatiotemporal information-theoretic metrics to unify exploration of unknown regions and tracking of known targets through long-horizon continuous trajectory optimization. During the interception phase, the system seamlessly transitions to a pure pursuit controller to achieve physical capture. A carefully designed multi-objective cost function effectively balances search and tracking priorities, establishing a complete closed-loop pipeline from planning to execution. Simulation results demonstrate superior performance over existing baselines, and real-world field experiments with an autonomous surface vehicle (ASV) in open water validate the system’s effectiveness and practical feasibility.

This paper addresses the challenge of autonomous aerial interception of non-cooperative, highly maneuverable unmanned aerial vehicles (UAVs) with unknown trajectories. We propose a real-time, robust interception method tailored for onboard net-launching platforms. Our approach features: (1) a Fast-Response Proportional Navigation (FRPN) guidance law that significantly improves dynamic responsiveness and capture success rate; (2) a state estimation algorithm integrating Interactive Multiple Model (IMM) filtering with an adaptive novel measurement model, relaxing reliance on predefined motion models and enhancing tracking robustness against aggressive maneuvers; and (3) an end-to-end closed-loop autonomous control architecture. Extensive simulations—covering 100 complex trajectories equivalent to 14 flight hours—demonstrate FRPN’s superiority in both response speed and capture rate. Real-world flight experiments successfully intercept highly maneuvering targets, outperforming existing state-of-the-art methods.

This work proposes a method for inferring pursuer parameters and planning safe, time-optimal paths in a pursuit-evasion scenario with bounded turn rates. By deploying sacrificial agents that execute straight-line trajectories and observing binary outcomes—interception or survival—of their interactions with the adversary, the approach leverages both boundary and interior geometric reachable set models to inversely estimate pursuer dynamics. A custom loss function combined with multi-start gradient-based optimization enables robust parameter recovery, while Bayesian experimental design guided by the D-optimality criterion efficiently selects the most informative sacrificial trajectories. The method accurately identifies pursuer parameters within only 5–12 interactions. Leveraging these estimates, high-value agent trajectories are synthesized to simultaneously guarantee safety—by avoiding all feasible engagement zones—and achieve time optimality.

This work addresses the challenge that existing spacecraft trajectory optimization relies heavily on manual modeling, hindering safe and autonomous decision-making directly from high-level mission intent. To bridge this gap, the authors propose a novel autonomy framework that leverages behavioral sequences and waypoint constraints as intermediate abstractions to decouple high-level semantic reasoning from low-level safety-critical trajectory optimization. The approach employs a foundation model to generate intent-aligned behavioral plans, which are then refined into dynamically feasible trajectories through a waypoint generation module coupled with an optimization-based safety solver. Evaluated in close-proximity operations, the method achieves over 90% convergence rate using sequential convex programming (SCP) and yields a 1.5× improvement in the proportion of generated trajectories satisfying high-level intent specifications compared to heuristic baselines.

Addressing the lack of formal safety guarantees and dynamic adaptability for multi-vehicle formation flight in complex 3D urban air mobility (UAM) environments, this paper proposes a safety-critical cooperative framework based on a leader–follower architecture. The method integrates a precomputed backup trajectory set, an online formation tracking controller, and leader-path-dependent safe backup maneuvers—extending the Guardian algorithm to 3D multi-robot systems for the first time. All components are formally verified to ensure collision avoidance under bounded disturbances and actuation limits. The framework achieves provable safety without sacrificing real-time performance: in 100 randomized high-fidelity simulations, it attains 100% obstacle avoidance success, outperforming both Control Barrier Function (CBF) and Nonlinear Model Predictive Control (NMPC) baselines. Furthermore, experimental validation is conducted on a physical quadcopter swarm, confirming practical feasibility and robustness in real-world deployment.

This work addresses the problem of cooperative simultaneous interception of a stationary target under heterogeneous sensing topologies, where some interceptors lack seekers and the target state is only partially observable. A unified nonlinear estimation-guidance-control framework is proposed, wherein a prescribed-time distributed observer enables seekerless interceptors to estimate the target state. Cooperative guidance commands are generated by integrating an improved time-to-go estimation with a prescribed-time consensus protocol, and executed by canard-driven autopilots. The designed prescribed-time sliding mode control law guarantees nonsingular, full-chain convergence within a predetermined time. Simulations demonstrate that the approach achieves high-accuracy state estimation, rapid time consensus, and precise command tracking across diverse engagement geometries, effectively enabling multiple missiles to achieve simultaneous impact within a wide launch envelope.

Autonomous lane-following navigation for non-holonomic differential-drive robots operating without global localization or high-definition maps in dynamic, partially observable lane environments remains highly challenging. Method: This paper proposes a real-time end-to-end visual navigation framework integrating YOLOP-based multi-task perception, 2D-to-3D lane-line reconstruction, arc-length-uniform sampling, and robust cubic polynomial fitting via QR decomposition; it further introduces a Lyapunov-based nonlinear controller ensuring strict stability of the perception–planning–control loop. Results: The system runs in real time (>30 FPS) on embedded hardware, generates smooth trajectories, and guarantees asymptotic convergence of both lateral position and heading errors. Experimental evaluation demonstrates significantly enhanced robustness and adaptability in dynamic scenarios, marking the first end-to-end visual navigation framework with provable closed-loop stability under partial observability and motion constraints.