We study in this paper the path properties of the Brownian motion and super-Brownian motion on the fractal structure-the Sierpinski gasket. At first some results about the limiting behaviour of its increments are obta...We study in this paper the path properties of the Brownian motion and super-Brownian motion on the fractal structure-the Sierpinski gasket. At first some results about the limiting behaviour of its increments are obtained and a kind of law of iterated logarithm is proved. Then A Lower bound of the spreading speed of its corresponding super-Brownian motion is obtained.展开更多
为提升弹载成像制导中运动模糊图像目标检测的精确性与效率,提出一种轻量化且高效的运动模糊图像目标检测(Lighter and More Effective Motion-blurred Image Object Detection,LEMBD)网络。通过深入分析运动模糊图像的成因,基于成像机...为提升弹载成像制导中运动模糊图像目标检测的精确性与效率,提出一种轻量化且高效的运动模糊图像目标检测(Lighter and More Effective Motion-blurred Image Object Detection,LEMBD)网络。通过深入分析运动模糊图像的成因,基于成像机理构建了专用的运动模糊图像数据集。在不增加网络参数的前提下,采用共享权重的孪生网络设计,并引入先验知识,将清晰图像的特征学习用于模糊图像的特征提取,以同时实现对清晰与模糊图像的精准检测。此外,设计了部分深度可分离卷积替代普通卷积,显著减少了网络的参数量与计算量,并提升了学习性能。为进一步优化特征融合质量,提出跨层路径聚合特征金字塔网络,有效利用低级特征的细节信息和高级特征的语义信息。实验结果表明,所提LEMBD网络在运动模糊图像目标检测任务中的性能优于传统目标检测方法和主流运动模糊检测算法,能够为精确制导任务提供更精准的目标相对位置信息。展开更多
This paper conducts a series of case studies on a novel Simultaneous Path and Motion Planning (SiPaMoP) approach [1] to multiple autonomous or Automated Guided Vehicle (AGV) motion coordination in bidirectional networ...This paper conducts a series of case studies on a novel Simultaneous Path and Motion Planning (SiPaMoP) approach [1] to multiple autonomous or Automated Guided Vehicle (AGV) motion coordination in bidirectional networks. The SiPaMoP approach plans collision-free paths for vehicles based on the principle of shortest path by dynamically changing the vehicles’ paths,traveling speeds or waiting times,whichever gives the shortest traveling time. It integrates path planning,collision avoidance and motion planning into a comprehensive model and optimizes the vehicles’ path and motion to minimize the completion time of a set of tasks. Five case studies,i.e.,head-on collision avoidance,catching-up collision avoidance,buffer node generation and collision avoidance,prioritybased motion coordination,and safety distance based planning,are presented. The results demonstrated that the method can effectively plan the path and motion for a team of autonomous vehicles or AGVs,and solve the problems of traffic congestion and collision under various conditions.展开更多
Assembly path planning is a crucial problem in assembly related design and manufacturing processes. Sampling based motion planning algorithms are used for computational assembly path planning. However, the performance...Assembly path planning is a crucial problem in assembly related design and manufacturing processes. Sampling based motion planning algorithms are used for computational assembly path planning. However, the performance of such algorithms may degrade much in environments with complex product structure, narrow passages or other challenging scenarios. A computational path planner for automatic assembly path planning in complex 3D environments is presented. The global planning process is divided into three phases based on the environment and specific algorithms are proposed and utilized in each phase to solve the challenging issues. A novel ray test based stochastic collision detection method is proposed to evaluate the intersection between two polyhedral objects. This method avoids fake collisions in conventional methods and degrades the geometric constraint when a part has to be removed with surface contact with other parts. A refined history based rapidly-exploring random tree （RRT） algorithm which bias the growth of the tree based on its planning history is proposed and employed in the planning phase where the path is simple but the space is highly constrained. A novel adaptive RRT algorithm is developed for the path planning problem with challenging scenarios and uncertain environment. With extending values assigned on each tree node and extending schemes applied, the tree can adapts its growth to explore complex environments more efficiently. Experiments on the key algorithms are carried out and comparisons are made between the conventional path planning algorithms and the presented ones. The comparing results show that based on the proposed algorithms, the path planner can compute assembly path in challenging complex environments more efficiently and with higher success. This research provides the references to the study of computational assembly path planning under complex environments.展开更多
In this note, we study a discrete time approximation for the solution of a class of delayed stochastic differential equations driven by a fractional Brownian motion with Hurst parameter H ∈(1/2,1). In order to prove ...In this note, we study a discrete time approximation for the solution of a class of delayed stochastic differential equations driven by a fractional Brownian motion with Hurst parameter H ∈(1/2,1). In order to prove convergence, we use rough paths techniques. Theoretical bounds are established and numerical simulations are displayed.展开更多
It is well known that the sufficient family of time-optimal paths for both Dubins' as well as Reeds-Shepp' s car models consist of the concatenation of circular arcs with maximum curvature and straight line se...It is well known that the sufficient family of time-optimal paths for both Dubins' as well as Reeds-Shepp' s car models consist of the concatenation of circular arcs with maximum curvature and straight line segments, all tangentially connected. These time-optimal solutions suffer from some drawbacks. Their discontinuous curvature profile, together with the wear and impairment on the control equipment that the bang-bang solutions induce, calls for ' smoother' and more supple reference paths to follow. Avoiding the bang-bang solutions also raises the robustness with respect to any possible uncertainties. In this paper, our main tool for generating these “nearly time-optimal” , but nevertheless continuous-curvature paths, is to use the Pontryagin Maximum Principle (PMP) and make an appropriate and cunning choice of the Lagrangian function. Despite some rewarding simulation results, this concept turns out to be numerically divergent at some instances. Upon a more careful investigation, it can be concluded that the problem at hand is nearly singular. This is seen by applying the PMP to Dubins car and studying the corresponding two point boundary value problem, which turn out to be singular. Realizing this, one is able to contradict the widespread belief that all the information about the motion of a mobile platform lies in the initial values of the auxiliary variables associated with the PMP. Keywords Time-optimal paths - Motion planning - Optimal control - Pontryagin maximum principle - UGV展开更多
针对高速自动驾驶车辆实时高精度的运动控制问题,提出一种上层为基于路径点Cost的路径点筛选器与基于横纵向轮胎力分析的速度规划器、下层为基于线性时变动力学模型预测的路径跟踪控制器与速度控制器的两层架构,并引入最小均方(Least Me...针对高速自动驾驶车辆实时高精度的运动控制问题,提出一种上层为基于路径点Cost的路径点筛选器与基于横纵向轮胎力分析的速度规划器、下层为基于线性时变动力学模型预测的路径跟踪控制器与速度控制器的两层架构,并引入最小均方(Least Mean Square,LMS)自适应状态估计器提升系统的抗噪性。路径点筛选器提升运算速度并减少筛选过程中的关键信息损失,速度规划器在安全行驶前提下生成最优速度曲线。路径跟踪控制器考虑跟踪偏差软约束,提升跟踪效果。LMS状态估计器基于在线矫正的动力学模型,对横摆角速度与横向速度在线估计。搭建dSPACE-TX2硬件在环仿真环境,在高速公路工况及双移线工况下对比所提出方案与传统运动跟踪控制。半实物仿真结果表明,所提出的运动控制架构提升了抗噪性能与21%的跟踪精度,且满足50 Hz高频控制的要求。展开更多
The solutions of the following bilinear stochastic differential equation are studied [GRAPHICS] where A(t)(k), B-t are (deterministic) continuous matrix-valued functions of t and w(1) (t),..., w(m) (t) are m independe...The solutions of the following bilinear stochastic differential equation are studied [GRAPHICS] where A(t)(k), B-t are (deterministic) continuous matrix-valued functions of t and w(1) (t),..., w(m) (t) are m independent standard Brownian motions. Conditions are given such that the solution is positive if the initial condition is positive. The equation the most probable path must satisfy is also derived and applied to a mathematical finance problem.展开更多
文摘We study in this paper the path properties of the Brownian motion and super-Brownian motion on the fractal structure-the Sierpinski gasket. At first some results about the limiting behaviour of its increments are obtained and a kind of law of iterated logarithm is proved. Then A Lower bound of the spreading speed of its corresponding super-Brownian motion is obtained.
文摘为提升弹载成像制导中运动模糊图像目标检测的精确性与效率,提出一种轻量化且高效的运动模糊图像目标检测(Lighter and More Effective Motion-blurred Image Object Detection,LEMBD)网络。通过深入分析运动模糊图像的成因,基于成像机理构建了专用的运动模糊图像数据集。在不增加网络参数的前提下,采用共享权重的孪生网络设计,并引入先验知识,将清晰图像的特征学习用于模糊图像的特征提取,以同时实现对清晰与模糊图像的精准检测。此外,设计了部分深度可分离卷积替代普通卷积,显著减少了网络的参数量与计算量,并提升了学习性能。为进一步优化特征融合质量,提出跨层路径聚合特征金字塔网络,有效利用低级特征的细节信息和高级特征的语义信息。实验结果表明,所提LEMBD网络在运动模糊图像目标检测任务中的性能优于传统目标检测方法和主流运动模糊检测算法,能够为精确制导任务提供更精准的目标相对位置信息。
文摘This paper conducts a series of case studies on a novel Simultaneous Path and Motion Planning (SiPaMoP) approach [1] to multiple autonomous or Automated Guided Vehicle (AGV) motion coordination in bidirectional networks. The SiPaMoP approach plans collision-free paths for vehicles based on the principle of shortest path by dynamically changing the vehicles’ paths,traveling speeds or waiting times,whichever gives the shortest traveling time. It integrates path planning,collision avoidance and motion planning into a comprehensive model and optimizes the vehicles’ path and motion to minimize the completion time of a set of tasks. Five case studies,i.e.,head-on collision avoidance,catching-up collision avoidance,buffer node generation and collision avoidance,prioritybased motion coordination,and safety distance based planning,are presented. The results demonstrated that the method can effectively plan the path and motion for a team of autonomous vehicles or AGVs,and solve the problems of traffic congestion and collision under various conditions.
基金supported by National Natural Science Foundation of China(Grant No. 51275047)Fund of National Engineering and Research Center for Commercial Aircraft Manufacturing of China(Grant No. 07205)Specialized Research Fund for the Doctoral Program of Higher Education of China(Grant No. 20091101110010)
文摘Assembly path planning is a crucial problem in assembly related design and manufacturing processes. Sampling based motion planning algorithms are used for computational assembly path planning. However, the performance of such algorithms may degrade much in environments with complex product structure, narrow passages or other challenging scenarios. A computational path planner for automatic assembly path planning in complex 3D environments is presented. The global planning process is divided into three phases based on the environment and specific algorithms are proposed and utilized in each phase to solve the challenging issues. A novel ray test based stochastic collision detection method is proposed to evaluate the intersection between two polyhedral objects. This method avoids fake collisions in conventional methods and degrades the geometric constraint when a part has to be removed with surface contact with other parts. A refined history based rapidly-exploring random tree （RRT） algorithm which bias the growth of the tree based on its planning history is proposed and employed in the planning phase where the path is simple but the space is highly constrained. A novel adaptive RRT algorithm is developed for the path planning problem with challenging scenarios and uncertain environment. With extending values assigned on each tree node and extending schemes applied, the tree can adapts its growth to explore complex environments more efficiently. Experiments on the key algorithms are carried out and comparisons are made between the conventional path planning algorithms and the presented ones. The comparing results show that based on the proposed algorithms, the path planner can compute assembly path in challenging complex environments more efficiently and with higher success. This research provides the references to the study of computational assembly path planning under complex environments.
基金supported by MATH-AmSud 18-MATH-07 SaS MoTiDep ProjectHERMES project 41305+1 种基金partially supported by the Project ECOS-CONICYT C15E05,REDES 150038,MATH-AmSud 18-MATH-07 SaS MoTiDep Project and Fondecyt(1171335)supported by NSF(Grant DMS-1613163)
文摘In this note, we study a discrete time approximation for the solution of a class of delayed stochastic differential equations driven by a fractional Brownian motion with Hurst parameter H ∈(1/2,1). In order to prove convergence, we use rough paths techniques. Theoretical bounds are established and numerical simulations are displayed.
文摘It is well known that the sufficient family of time-optimal paths for both Dubins' as well as Reeds-Shepp' s car models consist of the concatenation of circular arcs with maximum curvature and straight line segments, all tangentially connected. These time-optimal solutions suffer from some drawbacks. Their discontinuous curvature profile, together with the wear and impairment on the control equipment that the bang-bang solutions induce, calls for ' smoother' and more supple reference paths to follow. Avoiding the bang-bang solutions also raises the robustness with respect to any possible uncertainties. In this paper, our main tool for generating these “nearly time-optimal” , but nevertheless continuous-curvature paths, is to use the Pontryagin Maximum Principle (PMP) and make an appropriate and cunning choice of the Lagrangian function. Despite some rewarding simulation results, this concept turns out to be numerically divergent at some instances. Upon a more careful investigation, it can be concluded that the problem at hand is nearly singular. This is seen by applying the PMP to Dubins car and studying the corresponding two point boundary value problem, which turn out to be singular. Realizing this, one is able to contradict the widespread belief that all the information about the motion of a mobile platform lies in the initial values of the auxiliary variables associated with the PMP. Keywords Time-optimal paths - Motion planning - Optimal control - Pontryagin maximum principle - UGV
文摘针对高速自动驾驶车辆实时高精度的运动控制问题,提出一种上层为基于路径点Cost的路径点筛选器与基于横纵向轮胎力分析的速度规划器、下层为基于线性时变动力学模型预测的路径跟踪控制器与速度控制器的两层架构,并引入最小均方(Least Mean Square,LMS)自适应状态估计器提升系统的抗噪性。路径点筛选器提升运算速度并减少筛选过程中的关键信息损失,速度规划器在安全行驶前提下生成最优速度曲线。路径跟踪控制器考虑跟踪偏差软约束,提升跟踪效果。LMS状态估计器基于在线矫正的动力学模型,对横摆角速度与横向速度在线估计。搭建dSPACE-TX2硬件在环仿真环境,在高速公路工况及双移线工况下对比所提出方案与传统运动跟踪控制。半实物仿真结果表明,所提出的运动控制架构提升了抗噪性能与21%的跟踪精度,且满足50 Hz高频控制的要求。
基金the General Research Fund of the University of Kansas.
文摘The solutions of the following bilinear stochastic differential equation are studied [GRAPHICS] where A(t)(k), B-t are (deterministic) continuous matrix-valued functions of t and w(1) (t),..., w(m) (t) are m independent standard Brownian motions. Conditions are given such that the solution is positive if the initial condition is positive. The equation the most probable path must satisfy is also derived and applied to a mathematical finance problem.
