基于抽象解释的迹划分技术研究

确保程序中没有运行时错误,对于软件安全性的保证十分重要。基于抽象解释的静态分析方法对程序语义进行抽象,是验证运行时错误最合适的形式化方法之一。然而抽象解释对于程序语义的抽象可能导致过近似问题,从而引发误报,降低了分析精度。因此提出了迹划分的技术,根据程序的迹对程序控制流图进行划分,对静态分析过程进行局部细化,减少了抽象解释过程中过近似引发的误报。迹划分技术以局部分析效率降低为代价换来了分析精度的提高。

面向区域目标的敏捷卫星非沿迹条带划分方法

考虑单星对区域目标的非沿迹条带划分问题,以最少划分条带数为主优化目标,以最小姿态机动幅度为次优化目标,建立了该问题模型;并提出了一种角度步进的旋转划分算法进行求解。问题求解过程中考虑了初始划分方向确定、条带分割与裁剪、观测顺序与观测时间的确定等多个子问题。在验证了该算法可行性基础上,设计3组实验,并制定任务完成率、任务观测时间、条带分割数量、总姿态机动幅度4个衡量指标对比了沿迹划分算法与此算法的划分效能,结果显示在对中、低纬目标的划分中,该算法划分效果显著。

基于船舶融合点迹行为识别的雷达监视系统

针对近海监控管理的需求,将电子海图、雷达监控、AIS数据与CCTV技术相结合,通过集成平台对数据的融合处理,实现海洋交通的宏观、动态、实时、立体化的综合智能监控。同时提出一种基于地域信息位置特征点提取(Regional Information Feature Points Extraction,RIFPE)的点迹段划分方法。以某雷达基站为实验点,对已有船只的各项数据运用向量自回归和因子分析进行建模得到区域划分后的轨迹段的轨迹阈值,基于k最近邻算法(k NN)得到对轨迹阈值训练后的结果,最终对测试集进行轨迹行为判别。

Multi-robot task allocation for exploration

The problem of allocating a number of exploration tasks to a team of mobile robots in dynamic environments was studied. The team mission is to visit several distributed targets. The path cost of target is proportional to the distance that a robot has to move to visit the target. The team objective is to minimize the average path cost of target over all targets. Finding an optimal allocation is strongly NP-hard. The proposed algorithm can produce a near-optimal solution to it. The allocation can be cast in terms of a multi-round single-item auction by which robots bid on targets. In each auction round, one target is assigned to a robot that produces the lowest path cost of the target. The allocated targets form a forest where each tree corresponds a robot’s exploring targets set. Each robot constructs an exploring path through depth-first search in its target tree. The time complexity of the proposed algorithm is polynomial. Simulation experiments show that the allocating method is valid.

Cooperative merging control strategy of connected and automated vehicles on highways

To improve traffic performance when on-ramp vehicles merge into the mainstream,a collaborative merging control strategy is proposed to determine the merging sequence and trajectory control of vehicles.Merging trajectory planning takes the minimization of vehicle acceleration as the optimization objective.Either the variational method or the quadratic programming method is utilized to determine arrival time,optimal time and control variables for each vehicle.As a supplement,the adaptive cruise control(ACC)model is used to calculate each control variable in each time interval on special occasions.Simulation results show that the cooperative merging control strategy outperforms the optimal control strategy.The root mean square(RMS)of acceleration and the root mean square error(RMSE)of time headway are significantly decreased,with the reductions up to 90.1%and 25.2%,respectively.Under the cooperative control strategy,the difference between the average speed and desired speed consistently approaches zero.In addition,few or no collisions occur.To conclude,the proposed strategy favours the improvements in passenger comfort,traffic efficiency,traffic stability and safety around highway on-ramps.

Kinematic Analysis and Trajectory Planning of J-Groove Welding Robot

This paper introduces the complexity and particularity of tube-sphere intersection weld(J-groove weld) and establishes the mathematical model of tube-sphere intersection trajectory.Based on the characteristics of J-groove welds,the computational process of welding gun orientation is first simplified.Then the kinematic algorithm of a welding robot is obtained according to screw theory and exponential product formula.Finally,Solidworks and SimMechanics are employed to simulate the kinematics of the welding robot,which proves the feasibility of the kinematic algorithm.

Intelligent computing budget allocation for on-road tra jectory planning based on candidate curves

In this paper, on-road trajectory planning is solved by introducing intelligent computing budget allocation(ICBA) into a candidate-curve-based planning algorithm, namely, ordinal-optimization-based differential evolution(OODE). The proposed algorithm is named IOODE with ‘I' representing ICBA. OODE plans the trajectory in two parts: trajectory curve and acceleration profile. The best trajectory curve is picked from a set of candidate curves, where each curve is evaluated by solving a subproblem with the differential evolution(DE) algorithm. The more iterations DE performs, the more accurate the evaluation will become. Thus, we intelligently allocate the iterations to individual curves so as to reduce the total number of iterations performed. Meanwhile, the selected best curve is ensured to be one of the truly top curves with a high enough probability. Simulation results show that IOODE is 20% faster than OODE while maintaining the same performance in terms of solution quality. The computing budget allocation framework presented in this paper can also be used to enhance the efficiency of other candidate-curve-based planning methods.