基于深度神经网络的空中目标作战意图识别

传统基于空中目标特征状态推理作战意图的方法,需要大量的领域专家知识对特征状态的权重、先验概率等进行量化,明确特征状态与意图之间的对应关系,而神经网络可以在领域专家知识不足条件下,通过自身训练得到特征状态与意图之间的规则。针对反向传播（BP）算法在更新网络节点权值时收敛速度慢、容易陷入局部最优的问题,通过引入ReLU（Rectified Linear Unit）激活函数和自适应矩估计（Adam）优化算法,设计了基于深度神经网络的作战意图识别模型,提高了模型收敛速度,有效地防止陷入局部最优。仿真结果表明,所提方法能够有效识别空中目标作战意图,获得更高的识别率。     

高职院校毕业生就业意向的调查与分析

对长沙地区与南京地区14所高职院校1400位毕业生的就业意向进行调查和分析,根据调查结论提出加强高职院校毕业生就业工作的对策和措施。     

Aerial-ground collaborative routing with time constraints

The advancement of autonomous technology makes electric-powered drones an excellent choice for flexible logistics services at the last mile delivery stage. To reach a balance between green transportation and competitive edge, the collaborative routing of drones in the air and trucks on the ground is increasingly invested in the next generation of delivery, where it is particularly reasonable to consider customer time windows and time-dependent travel times as two typical time-related factors in daily services. In this paper, we propose the Vehicle Routing Problem with Drones under Time constraints (VRPD-T) and focus on the time constraints involved in realistic scenarios during the delivery. A mixed-integer linear programming model has been developed to minimize the total delivery completion time. Furthermore, to overcome the limitations of standard solvers in handling large-scale complex issues, a space-time hybrid heuristic-based algorithm has been developed to effectively identify a high-quality solution. The numerical results produced from randomly generated instances demonstrate the effectiveness of the proposed algorithm.