论我国高校协同创新的改进与完善

高校协同创新是指高校与科研机构、企业、政府之间,围绕国家重大科技项目和各行业关键技术,投入各自优势资源,合作攻关,力求取得重大突破的创新性活动。高校协同创新具备动态性、多元性、融合性、持续性的特点。要改进和完善我国高校协同创新,需要创建良好的协同创新组织方式、培养协同创新的思想观念、形成良好的创新型人才培养机制。     

SLOWING DOWN TO KEEP THE LEAD IN MILITARY TECHNOLOGY

We develop a model of military technology competition among states. States can choose to introduce new military technology, mimic rivals’ level of technology, or withdraw from the contest. States can choose to implement any level of technology within their current feasible technologies. We find that states with significant technological leads should sometimes withhold new technologies, only strategically releasing them to trump rivals’ efforts. We develop the model by refining Admiral Jackie Fisher’s roughly articulated concept of ‘plunging’. We then use this refined argument to reanalyze the case of naval rivalries among European powers between the Crimean War and the First World War. Finally, we conclude by discussing the model’s implications for current US military force structure planning.     

U.S. policy on preventive war and preemption

In the aftermath of September 11, 2001 and Hurricane Katrina, the United States government has spent tens of billions of dollars to improve the nation's ability to respond to a natural disaster or terrorist attack, but the emphasis on immediate first response has left many long-term environmental, political, and technological challenges unaddressed. Although a dirty bomb attack is unlikely to yield the same amount of physical devastation and death as caused by Katrina or a nuclear weapon, the social, psychological, and economic impact would be enormous. At present, however, the United States lacks the technology necessary to decontaminate a large, densely populated urban area under time, political, and economic constraints. This article reviews past cleanup experiences and current decontamination capabilities to consider the long-term implications of a dirty bomb, identifies weaknesses in America's existing response capabilities, notes possible areas of political friction, and considers the implications of the failure to adequately prepare. Having the appropriate decontamination techniques established and long-term plans in place before an incident occurs will significantly improve the government's ability to protect public and environmental security, establish a viable decontamination strategy, allow residents to return to their homes, and get the local economy back on its feet.     

Generalized orienteering problem with resource dependent rewards

We introduce a generalized orienteering problem (OP) where, as usual, a vehicle is routed from a prescribed start node, through a directed network, to a prescribed destination node, collecting rewards at each node visited, to maximize the total reward along the path. In our generalization, transit on arcs in the network and reward collection at nodes both consume a variable amount of the same limited resource. We exploit this resource trade‐off through a specialized branch‐and‐bound algorithm that relies on partial path relaxation problems that often yield tight bounds and lead to substantial pruning in the enumeration tree. We present the smuggler search problem (SSP) as an important real‐world application of our generalized OP. Numerical results show that our algorithm applied to the SSP outperforms standard mixed‐integer nonlinear programming solvers for moderate to large problem instances. We demonstrate model enhancements that allow practitioners to represent realistic search planning scenarios by accounting for multiple heterogeneous searchers and complex smuggler motion. © 2013 Wiley Periodicals, Inc. Naval Research Logistics, 2013     

舰艇防空火力规划与调度方法研究

根据舰艇防空武器拦截过程的时间连续性和拦截效果的不确定性特点,提出了智能规划和优化调度相结合的舰艇防空计划生成方法。在规划阶段,先用智能规划描述语言PDDL对舰艇防空问题进行建模;针对新的来袭目标或观察的拦截结果,利用智能规划的高效行动推理能力,快速生成可能的备选行动方案;在优化调度阶段,对已生成的备选方案的行动系列进行时间、资源冲突处理,并计算选择出当前最佳拦截方案。与现有的采用离散化处理的武器-目标分配方法相比,该规划与调度过程充分体现了拦截过程的连续性,拦截效果的不确定性,拦截过程中的武器协同,以及最佳拦截时机的选择等,更好地描述了舰艇防空的实际过程。最后,通过蒙特卡罗仿真试验,验证了算法的有效性,为舰艇的自动化防空提供了一种新的解决办法。     

面向武器装备体系组合规划的集成决策优化框架

针对当前武器装备体系组合规划存在选择空间规模大、决策目标数量多等问题,提出一种集成决策优化框架,用于组合选择和规划武器装备的发展型号、时间和数量。首先对武器装备体系组合规划问题的NP-Hard和高维多目标性质进行定量化分析和公式化描述;然后采用目标规划方法将该问题构建为双目标优化模型;再基于NSGA-II多目标演化计算方法,开发面向本问题的优化算法,求得该模型的Pareto解集合;最后通过TOPSIS方法,从Pareto解集合中求取符合决策者偏好的满意解。通过某侦察预警监视体系发展规划示例,验证了当给定经验数据和决策者偏好信息后,该框架可获得符合要求的武器装备体系组合规划方案,能够支撑武器装备体系发展论证和规划。     

基于压缩感知原理的融合判别信息的协作表示方法

提出了一种用于视觉分类任务的低计算复杂度且有效的图像表示方法。把协作表示和判别信息结合在统一框架内,是基于协作表示分类方法的一种扩展形式。测试样本的协作表示系数是稀疏的,这种基于冗余和过完备的表示对于遮挡和伪装而言是鲁棒的;此外,通过最小化类内散布矩阵和最大化类间散布矩阵的判别信息的挖掘,对于视觉分类问题也是很有帮助的。在一些基准数据库上的实验表明,提出的方法相对于现有的方法而言能够获得更有竞争力的表现。     

基于协同进化的多平台联合对地观测优化调度

在分析了卫星与无人机在执行观测与资源调度上的特性差异基础上,建立了多平台联合对地观测调度问题的数学模型,提出了多平台协同进化调度算法(MPCCPSA)进行求解。MPCCPSA采用分层式协同进化架构解决了不同类型观测方案统一调度生成问题。根据不同类型平台使用特性以及观测目标集合特点,采用分治-合作策略将其分解分配到各平台,顶层的交叉、变异操作保证各种群的多样性,底层的分治、合作算子保证卫星与无人机之间保持观测能力动态互补,在确保可行解的前提下加快收敛速度。仿真实验表明该方法能够有效解决空-天基多类型平台联合观测优化调度问题。     

基于进化算法的雷达对抗侦察无人机航路规划

雷达对抗侦察无人机在执行任务前要根据战场环境预先规划好最优路径。全面考虑战场环境威胁约束、机动性能约束和侦察任务约束,综合轮盘法、最优保存策略和6种进化算子形成进化算法,并对航路规划问题进行求解。仿真结果表明,利用约束量化方法和进化算法对雷达对抗侦察无人机进行航路规划是可行的。     

On valuing appreciating human assets in services

This paper derives optimal policies for when to hire, train, and lay off employees in service organizations when worker productivity appreciates deterministically or stochastically under constant, monotonic, and cyclic service demand. The model, an extension of the machine replacement problem from the engineering economy literature, uses infinite horizon linear programming to produce optimal personnel planning strategies and dual prices that characterize the workforce's economic value. The effect of employee attrition is also considered. In its valuation of human assets, unlike pure accounting approaches that only try to measure, this paper proposes an approach to capture the interaction between firm decisions pertaining to human resources and human resource value. © 2006 Wiley Periodicals, Inc. Naval Research Logistics, 2007