Active tracking control is essential for UAVs to perform autonomous operations in GPS-denied environments. In the active tracking task, UAVs take high-dimensional raw images as input and execute motor actions to actively follow the dynamic target. Most research focuses on three-stage methods, which entail perception first, followed by high-level decision-making based on extracted spatial information of the dynamic target, and then UAV movement control, using a low-level dynamic controller. Perception methods based on deep neural networks are powerful but require considerable effort for manual ground truth labeling. Instead, we unify the perception and decision-making stages using a high-level controller and then leverage deep reinforcement learning to learn the mapping from raw images to the high-level action commands in the V-REP-based environment, where simulation data are infinite and inexpensive. This end-to-end method also has the advantages of a small parameter size and reduced effort requirements for parameter turning in the decision-making stage. The high-level controller, which has a novel architecture, explicitly encodes the spatial and temporal features of the dynamic target. Auxiliary segmentation and motion-in-depth losses are introduced to generate denser training signals for the high-level controller’s fast and stable training. The high-level controller and a conventional low-level PID controller constitute our hierarchical active tracking control framework for the UAVs’ active tracking task. Simulation experiments show that our controller trained with several augmentation techniques sufficiently generalizes dynamic targets with random appearances and velocities, and achieves significantly better performance, compared with three-stage methods.

No abstract

Purpose By reducing the coating thickness of the weak scattering source, the coating weight of the absorbing material can be reduced by 35% with little effect on the RCS. Design/methodology/approach To alleviate the weight-increasing problem caused by a large number of coating of absorbing materials, a method for zonal coating of absorbing materials for a stealth helicopter was proposed. By appropriately reducing the thickness of the coating at the secondary scattering locations, the amount of coating used is significantly reduced. Findings Compared with the full-coated, the zonal coating scheme achieves the corresponding RCS reduction effect. Practical implications Zonal coating design can achieve the effect of reducing coating weight and cost. Originality/value The effects of different coating methods on RCS were verified by electromagnetic scattering simulation, and the applicability of the zonal coating design of the absorbing material to the stealth helicopter was verified.

The development of quantum radar technology presents a challenge to stealth targets, so it is necessary to study the quantum detection probability. In this study, an analytical expression of the quantum radar cross section (QRCS) for complex targets is presented. Based on this QRCS expression, a calculation method for the detection probability for quantum radar is creatively proposed. Moreover, a self-designed flying-wing stealth aircraft is adopted to obtain the detection probability distributions of the conventional radar and the quantum radar in different directions. As revealed by the result of this study, the detection probabilities of the quantum radar and the conventional radar are significantly different, and the detection probability of the quantum radar has obvious advantages in most regions with a certain distance.

Quantum radar is a novel detection method that combines radar and quantum technologies. It exceeds the detection threshold and poses a threat to conventional stealth targets. This work aims to derive the expression of the quantum radar cross-section of a new complex target. The calculation formula of QRCS was derived after introducing the relative photon parameters and vector dot product. Subsequently, a comprehensive optimization model of quantum stealth and lift–drag ratio based on a genetic algorithm was proposed for the waverider warhead. During the optimization process, we proposed an optimization method with the objective function of the QRCS pioneering design value and achieved better outcomes than the optimization method using the average value in terms of QRCS performance and lift–drag ratio in the important azimuths of the waverider. By changing the design variables of the waverider warhead and using this new optimization method, the QRCS of the waverider in the forward and lateral angles were minimized, remarkably improving the aerodynamic performance of the waverider. Similarly, the optimization results show that the proposed design value optimization method is feasible.