技术领域technical field
本发明涉及智能汽车技术领域，具体涉及一种基于人工势场法的无人驾驶车辆路径规划方法。The invention relates to the technical field of smart cars, in particular to a path planning method for an unmanned vehicle based on an artificial potential field method.
背景技术Background technique
路径规划是无人驾驶车辆研究领域的一项核心技术，它是指无人驾驶车辆根据多种传感器探测出的行驶环境信息，规划出一条从起始点到目标点的无碰撞路线，实现路径的最优化。路径规划主要包含两个步骤：一是建立包含障碍区域与自由区域的环境地图，二是在环境地图中选择合适的路径搜索算法，快速实时地搜索可行路径。路径规划结果对车辆行驶起着导航作用。它引导车辆从当前位置行驶到达目标位置。目前在无人驾驶中运用的路径规划算法有很多，如遗传算法、粒子群算法、A*算法、RRT算法和人工势场算法等。其中，人工势场法具有计算量小、规划路径平滑和便于实时控制等优点而被广泛运用于机器人导航和避碰中。Path planning is a core technology in the research field of unmanned vehicles. It means that unmanned vehicles plan a collision-free route from the starting point to the target point according to the driving environment information detected by various sensors, and realize the path planning. optimize. Path planning mainly includes two steps: one is to establish an environment map including obstacle areas and free areas, and the other is to select an appropriate path search algorithm in the environment map to search for feasible paths quickly and in real time. The result of path planning plays a role of navigation for vehicle driving. It guides the vehicle to travel from the current location to the target location. There are many path planning algorithms currently used in unmanned driving, such as genetic algorithm, particle swarm algorithm, A* algorithm, RRT algorithm and artificial potential field algorithm. Among them, the artificial potential field method is widely used in robot navigation and collision avoidance due to its advantages of small amount of calculation, smooth planning path and easy real-time control.
人工势场法于1986年由Khatib提出。其基本思想是，假设在汽车行驶道路中存在一种虚拟的势场力，目标点对主车产生引力，障碍物对主车产生斥力，主车在引力和斥力的合力控制作用下由高势场向低势场移动，最终到达目标位置。The artificial potential field method was proposed by Khatib in 1986. The basic idea is, assuming that there is a virtual potential field force on the road where the car is driving, the target point produces gravitational force on the main vehicle, and the obstacle produces repulsive force on the main vehicle. The field moves towards the lower potential field and finally reaches the target position.
但是，传统人工势场法仍然存在不足之处，当车辆在行进过程中有可能出现所受的斥力和引力大小相等方向相反，此时车辆在人工势场中所受到的合力为零，车辆陷入局部最优解，从而无法到达目标点。当目标点附近存在障碍物时，随着无人驾驶车辆不断地向目标点逼近，就可能出现障碍物对车辆的斥力远远大于目标点对车辆的引力，从而使得车辆在目标点附近徘徊，出现目标不可达的问题。However, the traditional artificial potential field method still has shortcomings. When the vehicle is moving, the repulsive force and the attractive force may be equal in magnitude and opposite in direction. At this time, the resultant force on the vehicle in the artificial potential field is zero, and the vehicle is trapped local optimal solution, so that the target point cannot be reached. When there is an obstacle near the target point, as the unmanned vehicle continues to approach the target point, the repulsion of the obstacle to the vehicle may be much greater than the gravitational force of the target point to the vehicle, so that the vehicle hovers near the target point. There is a problem that the target cannot be reached.
因此，传统人工势场法在复杂场景中存在易陷入局部最小点和目标不可达的问题，为解决这些问题，国内外学者对其进行了相应的改进研究：其一是通过在人工势场法的斥力场函数中加入汽车与目标点间距离的方法，使汽车只有在到达目标点时，斥力和引力才同时为零，改进后的人工势场法可以在静态环境中为车辆规划出安全的避障路径；其二是使用选取在障碍物一侧的中间目标点代替真正的目标点引导机器人，使机器人摆脱局部极小点；其三是通过分别引入RRT算法来弥补传统人工势场法容易形成局部最优解的不足；其四是分别通过引入斥力系数和调节因子以及道路边界斥力场模型，设立虚拟局部目标点，建立改进的路径规划模型从而有效地实现智能车辆的避撞局部路径；其五是通过优化吸引力场和排斥力场，并提出一种势场填充策略，使得移动机器人可以找到一条更好、无碰撞的目标路径。但是，上述改进仍然存在无法正常到达的问题。Therefore, the traditional artificial potential field method has the problem that it is easy to fall into the local minimum point and the target is unreachable in complex scenes. The method of adding the distance between the car and the target point to the repulsive force field function, so that the repulsive force and the gravitational force are zero at the same time only when the car reaches the target point. The second is to use the intermediate target point selected on the side of the obstacle instead of the real target point to guide the robot, so that the robot can get rid of the local minimum point; the third is to make up for the difficulty of the traditional artificial potential field method by introducing the RRT algorithm respectively. The lack of forming a local optimal solution; the fourth is to establish a virtual local target point by introducing the repulsion coefficient and adjustment factor and the road boundary repulsion field model, and establish an improved path planning model to effectively realize the collision avoidance local path of the intelligent vehicle; The fifth is to optimize the attractive field and repulsive force field, and propose a potential field filling strategy, so that the mobile robot can find a better, collision-free target path. However, the above-mentioned improvements still have the problem that they cannot be reached normally.
发明内容Contents of the invention
本发明的发明目的是，针对上述问题，提供一种基于人工势场法的无人驾驶车辆路径规划方法，用以解决无人驾驶车辆存在易陷入局部最小点和目标不可达的问题。The object of the present invention is to address the above problems and provide a path planning method for unmanned vehicles based on the artificial potential field method to solve the problems that unmanned vehicles are prone to fall into local minimum points and unreachable targets.
为达到上述目的，本发明所采用的技术方案是：一种基于人工势场法的无人驾驶车辆路径规划方法，包括以下步骤：In order to achieve the above object, the technical solution adopted in the present invention is: a method for path planning of unmanned vehicles based on the artificial potential field method, comprising the following steps:
1)、构建无人驾驶车行驶的二维空间模型，于所述二维空间模型内对无人驾驶车起始点、障碍物以及目标点的坐标进行定位，确定障碍物的个数n，确定无人驾驶车行驶的步长l；1), construct the two-dimensional space model of unmanned vehicle driving, locate the coordinates of unmanned vehicle starting point, obstacle and target point in described two-dimensional space model, determine the number n of obstacles, determine The step length l of the driverless car;
2)、建立由所述障碍物对无人驾驶车产生的斥力场和所述目标点对车辆产生的引力场叠加而成的虚拟势场，所述虚拟势场对所述无人驾驶车辆产生的作用力引导着无人驾驶车辆朝目标行进；2), establishing a virtual potential field superimposed by the repulsion field produced by the obstacle on the unmanned vehicle and the gravitational field produced by the target point on the vehicle, and the virtual potential field produces The force of the driver guides the unmanned vehicle towards the target;
3)使无人驾驶车行驶一个单位步长l，判断无人驾驶车是否陷入局部最小值点，如果是则调用步骤4)，否则进行步骤5)；3) Make the unmanned vehicle drive a unit step length l, judge whether the unmanned vehicle is trapped in a local minimum point, if so, call step 4), otherwise proceed to step 5);
4)改变斥力在X轴上的分量后回到步骤2)重新开始；4) After changing the component of the repulsive force on the X-axis, return to step 2) and start again;
5)判断无人驾驶车是否行进到目标点附近的影响距离内造成目标不可达，如果是则调用步骤6)，否则进行步骤7)；5) Judging whether the unmanned vehicle travels to the influence distance near the target point and causes the target to be unreachable, if so, call step 6), otherwise proceed to step 7);
6)在斥力势场函数中引入安全距离ρs和无人驾驶车到目标点之间的距离ρt后回到步骤(2)重新开始；6) Introduce the safety distance ρs and the distance ρt between the unmanned vehicle and the target point in the repulsion potential field function, and return to step (2) to start again;
7)判断无人驾驶车车是否到达目标点，如果是则停止路径规划画出路径，否则回到步骤(2)重新开始。7) Determine whether the unmanned vehicle has reached the target point, if so, stop path planning and draw the path, otherwise return to step (2) and start again.
作为一种改进的方式，步骤4)中，所述斥力在X轴上添加的变量系数为k，所述斥力在X轴上的分量为：As an improved way, in step 4), the variable coefficient added on the X-axis of the repulsive force is k, and the component of the repulsive force on the X-axis is:
，，Frep(X)为原所受斥力的合力大小,，ρ(X，Xo)为被控对象到障碍物之间的距离，ρo为障碍物的最大影响距离；,, Frep (X) is the resultant force of the original repulsive force, ρ(X, Xo ) is the distance between the controlled object and the obstacle, and ρo is the maximum influence distance of the obstacle;
所述斥力在Y轴上添加的变量系数为δ，所述斥力场在Y轴上的分量为：The variable coefficient added by the repulsive force on the Y axis is δ, and the component of the repulsive force field on the Y axis is:
作为一种改进的方式，步骤6)中，As an improved way, in step 6),
所述斥力势场函数为：The repulsion potential field function is:
其中，X＝(x,y)为无人驾驶车辆的当前位置坐标，Xo＝(xo，yo)为障碍物坐标，ρ(X，Xo)为被控对象到障碍物之间的距离，ρo为障碍物的最大影响距离，ρs为安全距离，ρt为无人驾驶车到目标点之间的距离，γ为安全距离所对应的斥力增益系数，β为未考虑添加安全距离时斥力增益系数。Among them, X = (x, y) is the current position coordinates of the unmanned vehicle, Xo = (xo , yo ) is the obstacle coordinates, ρ (X, Xo ) is the distance between the controlled object and the obstacle ρo is the maximum influence distance of obstacles, ρs is the safety distance, ρt is the distance between the unmanned vehicle and the target point, γ is the repulsion gain coefficient corresponding to the safety distance, and β is the Repulsion gain coefficient at safe distance.
其中，Frep1和Frep3的方向从障碍物指向无人驾驶车辆，Frep2和Frep4的方向从无人驾驶车辆指向目标点,Among them, the directions of Frep1 and Frep3 point to the unmanned vehicle from the obstacle, and the directions of Frep2 and Frep4 point to the target point from the unmanned vehicle,
当ρ(X，Xo)≤ρs时，When ρ(X, Xo )≤ρs ,
当ρ(X，Xo)≤ρo且ρ(X，Xo)＞ρs时，When ρ(X, Xo )≤ρo and ρ(X, Xo )>ρs ,
作为一种改进的方式，步骤2)中，所述引力场函数为；As an improved way, in step 2), the gravitational field function is;
其中，α为引力增益系数，所述引力函数为：Wherein, α is the gravitational gain coefficient, and the gravitational function is:
作为一种改进的方式，所述二维空间模型为包含障碍区域与自由区域的环境地图。As an improved manner, the two-dimensional space model is an environment map including obstacle areas and free areas.
作为一种改进的方式，目标点坐标为Xt＝(xt，yt)，被控对象到目标点之间的距离ρt为As an improved way, the coordinates of the target point are Xt = (xt , yt ), and the distance ρt between the controlled object and the target point is
由于采用上述技术方案，本发明具有以下有益效果：Owing to adopting above-mentioned technical scheme, the present invention has following beneficial effect:
1、本申请主要针对传统人工势场法存在的不足进行改进，当无人驾驶车辆在行驶过程中陷入局部最小值点时，通过在斥力函数的分量上添加变量系数改变斥力的方向，从而改变合力方向，使得无人驾驶车跳出局部最小值点，继续向目标点行进。1. This application mainly aims to improve the shortcomings of the traditional artificial potential field method. When the unmanned vehicle falls into a local minimum point during driving, the direction of the repulsive force is changed by adding variable coefficients to the components of the repulsive force function, thereby changing the The direction of the resultant force makes the unmanned vehicle jump out of the local minimum point and continue to move towards the target point.
2、当无人驾驶车行驶到目标点附近障碍物的影响距离范围内造成目标不可达问题时，通过加入无人驾驶车和目标点之间距离来提供新的斥力函数，并且添加安全距离来保证无人驾驶车行驶过程中的安全性。2. When the unmanned vehicle travels within the influence distance of the obstacle near the target point and the target is unreachable, a new repulsion function is provided by adding the distance between the unmanned vehicle and the target point, and a safety distance is added to Ensure the safety of unmanned vehicles during driving.
3、本申请将目标点与车辆之间的距离因子引入到斥力函数中，并且在障碍物的影响范围内添加了安全距离，使得在车辆行驶到障碍物的安全距离内时所受到的斥力大于未添加安全距离之前的斥力，从而保障了车辆行驶过程中的安全性。3. This application introduces the distance factor between the target point and the vehicle into the repulsion function, and adds a safety distance within the influence range of the obstacle, so that when the vehicle travels within the safety distance of the obstacle, the repulsion force it receives is greater than The repulsion before the safety distance is not added, thus ensuring the safety of the vehicle during driving.
由于步骤4)中，所述斥力在X轴上添加的变量系数为k，所述斥力在X轴上的分量为：Because in step 4), the variable coefficient added on the X-axis of the repulsive force is k, and the component of the repulsive force on the X-axis is:
所述斥力在Y轴上添加的变量系数为δ，所述斥力场在Y轴上的分量为：The variable coefficient added by the repulsive force on the Y axis is δ, and the component of the repulsive force field on the Y axis is:
，当合力为零时容易出现局部极小值点，车辆将会来回运动，因此对斥力在X轴和Y轴上的分量分别进行改进，使得合力大于零以避免这些情况的出现，这样车辆才可以顺利到达目标位置。, when the resultant force is zero, it is easy to appear a local minimum point, and the vehicle will move back and forth, so the components of the repulsive force on the X-axis and Y-axis are respectively improved so that the resultant force is greater than zero to avoid these situations, so that the vehicle can Can successfully reach the target location.
由于步骤6)中，所述斥力势场函数为：Because in step 6), described repulsion potential field function is:
其中，X＝(x,y)为无人驾驶车辆的当前位置坐标，Xo＝(xo，yo)为障碍物坐标，ρ(X，Xo)为被控对象到障碍物之间的距离，ρo为障碍物的最大影响距离，ρs为安全距离，ρt为无人驾驶车到目标点之间的距离，γ为安全距离所对应的斥力增益系数，β为未考虑添加安全距离时斥力增益系数，当车辆行驶到目标点附近的障碍物影响范围内时，可能造成斥力大于引力，使得车辆在目标点附近徘徊，无法停止运动。因此，将目标点与车辆之间的距离因子引入到斥力函数中，并且在障碍物的影响范围内添加了安全距离，使得在车辆行驶到障碍物的安全距离内时所受到的斥力大于未添加安全距离之前的斥力，从而保障了车辆行驶过程中的安全性。Among them, X = (x, y) is the current position coordinates of the unmanned vehicle, Xo = (xo , yo ) is the obstacle coordinates, ρ (X, Xo ) is the distance between the controlled object and the obstacle ρo is the maximum influence distance of obstacles, ρs is the safety distance, ρt is the distance between the unmanned vehicle and the target point, γ is the repulsion gain coefficient corresponding to the safety distance, and β is the The repulsion gain coefficient is the safe distance. When the vehicle travels within the influence range of obstacles near the target point, the repulsion force may be greater than the gravitational force, making the vehicle hover near the target point and cannot stop moving. Therefore, the distance factor between the target point and the vehicle is introduced into the repulsion function, and a safety distance is added within the influence range of the obstacle, so that when the vehicle travels within the safety distance of the obstacle, the repulsion force it receives is greater than that without The repulsive force before the safe distance ensures the safety of the vehicle during driving.
附图说明Description of drawings
图1是改进前斥力势场强度立体图；Fig. 1 is a three-dimensional view of the strength of the repulsion potential field before improvement;
图2是改进后斥力势场强度立体图；Fig. 2 is a three-dimensional view of the strength of the repulsion potential field after improvement;
图3是改进前斥力势场强度侧面图；Fig. 3 is a side view of the strength of the repulsion potential field before improvement;
图4是改进后斥力势场强度侧面图；Figure 4 is a side view of the strength of the repulsion potential field after improvement;
图5是改进前总势场强度平面图；Fig. 5 is a plan view of total potential field strength before improvement;
图6是改进后总势场强度平面图；Fig. 6 is a plan view of the total potential field strength after improvement;
图7是改进前总势场强度侧面图；Fig. 7 is a side view of the total potential field strength before improvement;
图8是改进后总势场强度侧面图；Figure 8 is a side view of the improved total potential field strength;
图9是车辆陷入局部最小值点状态图；Fig. 9 is a state diagram of a vehicle trapped in a local minimum point;
图10是车辆处于目标不可达状态图；Fig. 10 is a diagram showing that the vehicle is in a target unreachable state;
图11是车辆跳出局部最小值点路径图；Fig. 11 is a path diagram of a vehicle jumping out of a local minimum point;
图12是车辆目标可达的路径图；Fig. 12 is a path diagram that the vehicle target can reach;
图13是传统方法与改进方法的目标可达路径对比图。Figure 13 is a comparison diagram of the target reachable path between the traditional method and the improved method.
具体实施方式Detailed ways
本发明公开了一种基于人工势场法的无人驾驶车辆路径规划方法，包括以下步骤：The invention discloses a path planning method for an unmanned vehicle based on an artificial potential field method, comprising the following steps:
1)、构建无人驾驶车行驶的二维空间模型，二维空间模型为包含障碍区域与自由区域的环境地图，确定障碍物的个数n，于二维空间模型内对无人驾驶车起始点、障碍物以及目标点的坐标进行定位，目标点坐标为Xt＝(xt，yt)，被控对象(即无人驾驶车)到目标点之间的距离ρt为：确定无人驾驶车行驶的步长l。1) Construct a two-dimensional space model for the driving of the unmanned vehicle. The two-dimensional space model is an environmental map including obstacle areas and free areas. Determine the number n of obstacles. The coordinates of the starting point, the obstacle and the target point are used for positioning. The coordinates of the target point are Xt = (xt , yt ), and the distance ρt between the controlled object (ie unmanned vehicle) and the target point is: Determine the step size l of the driverless car.
2)、建立由障碍物对无人驾驶车产生的斥力场和目标点对车辆产生的引力场叠加而成的虚拟势场，其中，引力场函数为；α为引力增益系数，引力函数为：虚拟势场对无人驾驶车辆产生的作用力引导着无人驾驶车辆朝目标行进。2) Establish a virtual potential field formed by superimposing the repulsive field generated by obstacles on the unmanned vehicle and the gravitational field generated by the target point on the vehicle, wherein the gravitational field function is; α is the gravitational gain coefficient, and the gravitational function is: The force generated by the virtual potential field on the unmanned vehicle guides the unmanned vehicle towards the target.
3)无人驾驶车行驶一个单位步长l，然后判断是否陷入局部最小值点，如果是则调用步骤4)，否则进行步骤5)。3) The unmanned vehicle travels for a unit step length l, and then judges whether it is trapped in a local minimum point, if so, call step 4), otherwise, go to step 5).
4)改变斥力在X轴上的分量后回到步骤2)重新开始，具体改变公式为：4) After changing the component of the repulsive force on the X-axis, return to step 2) and start again. The specific change formula is:
斥力在X轴上添加的变量系数为k，斥力在X轴上的分量为：The variable coefficient added by the repulsion on the X-axis is k, and the component of the repulsion on the X-axis is:
，，Frep(X)为原所受斥力的合力大小,，ρ(X，Xo)为被控对象到障碍物之间的距离，ρo为障碍物的最大影响距离；,, Frep (X) is the resultant force of the original repulsive force, ρ(X, Xo ) is the distance between the controlled object and the obstacle, and ρo is the maximum influence distance of the obstacle;
斥力在Y轴上添加的变量系数为δ，斥力场在Y轴上的分量为：The variable coefficient added by the repulsive force on the Y axis is δ, and the component of the repulsive force field on the Y axis is:
，当合力为零时容易出现局部极小值点，车辆将会来回运动，因此使斥力在X轴和Y轴上的分量进行改进，使得合力大于零以避免这些情况的出现，这样车辆才可以顺利到达目标位置。, when the resultant force is zero, it is easy to appear a local minimum point, and the vehicle will move back and forth, so the components of the repulsive force on the X-axis and Y-axis are improved so that the resultant force is greater than zero to avoid these situations, so that the vehicle can successfully reached the target location.
5)判断无人驾驶车是否行进到目标点附近的影响距离内造成目标不可达，如果是则调用步骤6)，否则进行步骤7)；5) Judging whether the unmanned vehicle travels to the influence distance near the target point and causes the target to be unreachable, if so, call step 6), otherwise proceed to step 7);
6)在斥力势场函数中引入安全距离ρs和无人驾驶车到目标点之间的距离ρt后得到斥力势场函数为：6) After introducing the safety distance ρs and the distance ρt between the unmanned vehicle and the target point into the repulsion potential field function, the repulsion potential field function is:
其中，X＝(x,y)为无人驾驶车辆的当前位置坐标，Xo＝(xo，yo)为障碍物坐标，ρ(X，Xo)为被控对象到障碍物之间的距离，ρo为障碍物的最大影响距离，ρs为安全距离，ρt为无人驾驶车到目标点之间的距离，γ为安全距离所对应的斥力增益系数，β为未考虑添加安全距离时斥力增益系数，回到步骤(2)重新开始(当车辆行驶到目标点附近的障碍物影响范围内时，可能造成斥力大于引力，使得车辆在目标点附近徘徊，无法停止运动。因此，将目标点与车辆之间的距离因子引入到斥力函数中，并且在障碍物的影响范围内添加了安全距离，使得在车辆行驶到障碍物的安全距离内时所受到的斥力大于未添加安全距离之前的斥力，从而保障了车辆行驶过程中的安全性)。Among them, X = (x, y) is the current position coordinates of the unmanned vehicle, Xo = (xo , yo ) is the obstacle coordinates, ρ (X, Xo ) is the distance between the controlled object and the obstacle ρo is the maximum influence distance of obstacles, ρs is the safety distance, ρt is the distance between the unmanned vehicle and the target point, γ is the repulsion gain coefficient corresponding to the safety distance, and β is the The repulsion gain coefficient at the safe distance, return to step (2) and start again (when the vehicle travels within the influence range of obstacles near the target point, the repulsion force may be greater than the gravitational force, making the vehicle wander around the target point and cannot stop. Therefore , the distance factor between the target point and the vehicle is introduced into the repulsion function, and a safety distance is added within the influence range of the obstacle, so that when the vehicle travels within the safety distance of the obstacle, the repulsion force it receives is greater than that without safety The repulsion before the distance, thus ensuring the safety of the vehicle during driving).
7)判断无人驾驶车车是否到达目标点，如果是则停止路径规划画出路径，否则回到步骤(2)重新开始。7) Determine whether the unmanned vehicle has reached the target point, if so, stop path planning and draw the path, otherwise return to step (2) and start again.
其中，上述步骤6)中：斥力函数为：Wherein, in above-mentioned step 6): repulsion function is:
其中,Frep1和Frep3的方向从障碍物指向无人驾驶车辆，Frep2和Frep4的方向从无人驾驶车辆指向目标点,Among them, the directions of Frep1 and Frep3 point to the unmanned vehicle from the obstacle, and the directions of Frep2 and Frep4 point to the target point from the unmanned vehicle,
当ρ(X，Xo)≤ρs时，When ρ(X, Xo )≤ρs ,
当ρ(X，Xo)≤ρo且ρ(X，Xo)＞ρs时，When ρ(X, Xo )≤ρo and ρ(X, Xo )>ρs ,
图1和图2分别为加入安全距离前后的斥力势能强度立体图，图3和图4为加入安全距离前后的斥力势能强度侧面图，由图可知无人驾驶车的行驶空间为侧面图的横轴，障碍物的位置坐标分别为(1.5，1)、(5，6)、(9，5,5)，目标点位置坐标为(10,10)，纵轴代表斥力势能值。能够看出改进后的斥力势能值在障碍物附近有所上升，从而提高了车辆行驶的安全性。Figure 1 and Figure 2 are perspective views of the repulsive potential energy intensity before and after adding the safety distance, and Figure 3 and Figure 4 are side views of the repulsive potential energy intensity before and after adding the safety distance. , the position coordinates of obstacles are (1.5, 1), (5, 6), (9, 5, 5), respectively, and the position coordinates of the target point are (10, 10), and the vertical axis represents the repulsion potential energy value. It can be seen that the improved repulsion potential energy value increases near obstacles, thereby improving the safety of the vehicle.
图5和图6分别为加入安全距离前后总势能的平面图，此时横轴为Y轴，同样能够看出改进后的斥力势能值在障碍物附近有所上升。Figure 5 and Figure 6 are the plan views of the total potential energy before and after adding the safety distance, and the horizontal axis is the Y axis. It can also be seen that the improved repulsion potential energy value increases near the obstacle.
图7和图8分别为增加安全距离前后总势能的侧面图，能够看出改进前后目标点的势能为零，障碍物附近的势能出现突变，当无人驾驶车到达目标点附近时，势能值减小，从而也避免出现无人驾驶车辆到达目标点势能不为零的情况。Figure 7 and Figure 8 are the side views of the total potential energy before and after increasing the safety distance. It can be seen that the potential energy of the target point before and after the improvement is zero, and the potential energy near the obstacle has a sudden change. When the unmanned vehicle reaches the vicinity of the target point, the potential energy value , so as to avoid the situation that the potential energy of the unmanned vehicle reaching the target point is not zero.
为了验证本申请的基于人工势场法的无人驾驶车辆路径规划方法的效果，因此设计实验对其进行仿真分析，改进人工势场算法实验步骤如下：In order to verify the effect of the unmanned vehicle path planning method based on the artificial potential field method of the present application, an experiment is designed to simulate and analyze it, and the experimental steps for improving the artificial potential field algorithm are as follows:
(1)构建无人驾驶车的运行空间，确定无人驾驶车的起始点和目标点的位置、引力和斥力的增益系数α和β、障碍物的个数n、障碍物影响距离ρo、安全距离ρs以及无人驾驶车行驶的步长l。(1) Construct the operating space of the unmanned vehicle, determine the position of the starting point and the target point of the unmanned vehicle, the gain coefficients of attraction and repulsion α and β, the number of obstacles n, the obstacle influence distance ρo , The safety distance ρs and the step size l of the unmanned vehicle.
(2)建立虚拟势场并分别计算引力和斥力的大小。(2) Establish a virtual potential field and calculate the magnitude of the attractive and repulsive forces respectively.
(3)计算合力大小。(3) Calculate the resultant force.
(4)无人驾驶车向下一位置运动后判断是否陷入局部最小值点，如果是则调用步骤(5)，否则进行步骤(6)。(4) After the unmanned vehicle moves to the next position, it is judged whether it is trapped in a local minimum point, if so, call step (5), otherwise, go to step (6).
(5)改变斥力在X轴上的分量后回到步骤(2)重新开始。(5) Return to step (2) and start again after changing the component of the repulsive force on the X axis.
(6)无人驾驶车是否行进到目标点附近的影响距离内，造成目标不可达，如果是则调用步骤(7)，否则进行步骤(8)。(6) Whether the unmanned vehicle travels within the influence distance near the target point, causing the target to be unreachable, if so, call step (7), otherwise, go to step (8).
(7)在斥力函数中加入安全距离和无人驾驶车到目标点之间的距离后回到步骤(2)重新开始。(7) After adding the safety distance and the distance between the unmanned vehicle and the target point in the repulsion function, return to step (2) and start again.
(8)无人驾驶车是否到达目标点，如果是则停止路径规划画出路径，否则回到步骤(2)重新开始。(8) Whether the unmanned vehicle has reached the target point, if so, stop the path planning and draw the path, otherwise return to step (2) and start again.
根据以上的实验步骤在Matlab仿真平台上分别对改进前后的人工势场法进行仿真实验。选取引力增益系数为15，斥力增益系数为4，障碍物影响距离为2.5，安全距离为1，无人驾驶车的行驶步长为0.2，最大迭代次数为600，无人驾驶车的起始点位置坐标为(0,0)，目标点位置坐标为(10,10)，γ根据多次重复实验选取最优值。According to the above experimental steps, the artificial potential field method before and after improvement is simulated on the Matlab simulation platform. Select the gravitational gain coefficient to be 15, the repulsive force gain coefficient to be 4, the obstacle influence distance to be 2.5, the safety distance to be 1, the driving step of the unmanned vehicle to be 0.2, the maximum number of iterations to be 600, the starting point position of the unmanned vehicle The coordinates are (0,0), the coordinates of the target point are (10,10), and the optimal value of γ is selected according to repeated experiments.
首先，只运用传统人工势场法进行仿真。当无人驾驶车、障碍物以及目标点在同一条直线上时，无人驾驶车在靠近目标点的同时所受的引力在逐渐减小，而无人驾驶车受到的斥力逐渐增大，无人驾驶车在某一点处会由于受力平衡，无法躲避障碍物，陷入局部最小值点，无法到达目标点，如图9所示。当目标点存在障碍物且无人驾驶车行驶到障碍物的影响距离内时，无人驾驶车的斥力大于引力，从而使无人驾驶车不能到达目标点，如图10所示。First, only the traditional artificial potential field method is used for simulation. When the unmanned vehicle, the obstacle, and the target point are on the same straight line, the gravitational force on the unmanned vehicle gradually decreases as it approaches the target point, while the repulsion force on the unmanned vehicle gradually increases. At a certain point, due to the force balance, the human-driven car cannot avoid obstacles, falls into a local minimum point, and cannot reach the target point, as shown in Figure 9. When there is an obstacle at the target point and the unmanned vehicle travels within the influence distance of the obstacle, the repulsive force of the unmanned vehicle is greater than the gravitational force, so that the unmanned vehicle cannot reach the target point, as shown in Figure 10.
运用改进后的人工势场法进行仿真得到的结果如图11和12所示。图11是在图9中无人驾驶车陷入局部最小值点的情况下对斥力在X轴和Y轴上分量的改进，使得无人驾驶车跳出局部最小值点继续向目标点行进。图12是图10在无人驾驶车出现目标不可达的情况下对无人驾驶车的斥力添加调节因子和安全距离的改进，使得无人驾驶车在到达目标点时受力平衡。The simulation results obtained by using the improved artificial potential field method are shown in Figures 11 and 12. Fig. 11 shows the improvement of the components of the repulsive force on the X-axis and Y-axis when the unmanned vehicle falls into the local minimum point in Fig. 9, so that the unmanned vehicle jumps out of the local minimum point and continues to travel to the target point. Fig. 12 is an improvement of Fig. 10 by adding an adjustment factor and a safety distance to the repulsive force of the unmanned vehicle when the target is unreachable by the unmanned vehicle, so that the unmanned vehicle is balanced when it reaches the target point.
从仿真结果可以看出改进后的算法能够很好地解决小车陷入局部最小值点和目标不可达的情况并引导着小车行进到目标点，并通过图13中传统加距离因子改进方法与本文改进距离因子及增加安全距离的方法所形成的目标可达路径进行对比，从而验证了改进人工势场法的有效性。From the simulation results, it can be seen that the improved algorithm can well solve the situation that the car is trapped in the local minimum point and the target is unreachable, and guide the car to the target point, and through the traditional improved method of adding distance factor in Figure 13 and the improved method in this paper The distance factor and the target reachable path formed by the method of increasing the safety distance are compared, thus verifying the effectiveness of the improved artificial potential field method.
本申请介绍了传统人工势场法的基本原理，分析了传统人工势场法在路径规划中陷入局部极小点和目标不可达情况的原因，通过引入安全距离和调节因子来解决目标不可达问题的同时提高无人驾驶的安全性能，通过在斥力分量上增加可变系数的方式使无人驾驶车辆跳出局部最小值点。最后在Matlab仿真环境下验证了改进算法的有效性。This application introduces the basic principles of the traditional artificial potential field method, analyzes the reasons why the traditional artificial potential field method falls into local minimum points and unreachable targets in path planning, and solves the problem of unreachable targets by introducing safety distances and adjustment factors At the same time, the safety performance of unmanned driving can be improved, and the unmanned vehicle can jump out of the local minimum point by adding a variable coefficient to the repulsive force component. Finally, the effectiveness of the improved algorithm is verified in the Matlab simulation environment.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910943784.4ACN110567478B (en) | 2019-09-30 | 2019-09-30 | A Path Planning Method for Unmanned Vehicles Based on Artificial Potential Field Method |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910943784.4ACN110567478B (en) | 2019-09-30 | 2019-09-30 | A Path Planning Method for Unmanned Vehicles Based on Artificial Potential Field Method |
| Publication Number | Publication Date |
|---|---|
| CN110567478Atrue CN110567478A (en) | 2019-12-13 |
| CN110567478B CN110567478B (en) | 2023-06-30 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910943784.4AActiveCN110567478B (en) | 2019-09-30 | 2019-09-30 | A Path Planning Method for Unmanned Vehicles Based on Artificial Potential Field Method |
| Country | Link |
|---|---|
| CN (1) | CN110567478B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111123976A (en)* | 2019-12-24 | 2020-05-08 | 一飞智控(天津)科技有限公司 | Unmanned aerial vehicle cluster path planning processing method based on artificial potential field and unmanned aerial vehicle |
| CN111506083A (en)* | 2020-05-19 | 2020-08-07 | 上海应用技术大学 | Safety obstacle avoidance method for industrial robot based on artificial potential field method |
| CN112150634A (en)* | 2020-08-31 | 2020-12-29 | 浙江工业大学 | A large-scale virtual scene roaming method based on multi-person redirection |
| CN112923944A (en)* | 2021-01-29 | 2021-06-08 | 的卢技术有限公司 | Automatic driving path planning method and system and computer readable storage medium |
| CN113002537A (en)* | 2021-03-16 | 2021-06-22 | 镇江康飞汽车制造股份有限公司 | Vehicle active collision avoidance method based on artificial potential field method |
| CN113110441A (en)* | 2021-04-09 | 2021-07-13 | 江苏大学 | Agricultural unmanned vehicle cluster operation method based on ultra wide band |
| US20210272466A1 (en)* | 2020-02-28 | 2021-09-02 | Pablo Air Co., Ltd. | Method of avoiding collision of unmanned aerial vehicle |
| CN114077255A (en)* | 2021-11-22 | 2022-02-22 | 江苏理工学院 | A Way-finding Method for Intelligent Vehicles Based on Ellipse Model Artificial Potential Field Method |
| CN114265410A (en)* | 2021-12-25 | 2022-04-01 | 长安大学 | Local path planning method and system based on multi-computational power fusion |
| CN114442637A (en)* | 2022-02-10 | 2022-05-06 | 北京理工大学 | A local dynamic obstacle avoidance path planning method for unmanned vehicles |
| CN114460965A (en)* | 2022-01-21 | 2022-05-10 | 上海应用技术大学 | Unmanned aerial vehicle three-dimensional obstacle avoidance method based on improved artificial potential field method |
| CN115092183A (en)* | 2022-07-15 | 2022-09-23 | 东风柳州汽车有限公司 | Vehicle active obstacle avoidance control method and system based on potential field force |
| CN115416016A (en)* | 2022-08-15 | 2022-12-02 | 浙江工业大学 | Mechanical arm obstacle avoidance path planning method based on improved artificial potential field method |
| CN115793648A (en)* | 2022-11-29 | 2023-03-14 | 无锡学院 | An AGV path planning method |
| CN117369482A (en)* | 2023-12-06 | 2024-01-09 | 华润数字科技有限公司 | Path planning method, device and equipment for mobile robot and storage medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016045615A1 (en)* | 2014-09-25 | 2016-03-31 | 科沃斯机器人有限公司 | Robot static path planning method |
| CN105974917A (en)* | 2016-05-11 | 2016-09-28 | 江苏大学 | Vehicle obstacle-avoidance path planning research method based on novel manual potential field method |
| CN106843235A (en)* | 2017-03-31 | 2017-06-13 | 深圳市靖洲科技有限公司 | It is a kind of towards the Artificial Potential Field path planning without person bicycle |
| CN108469828A (en)* | 2018-03-23 | 2018-08-31 | 哈尔滨工程大学 | A kind of AUV Route planners improving artificial potential field optimization algorithm |
| CN109358637A (en)* | 2018-05-25 | 2019-02-19 | 武汉科技大学 | A three-dimensional obstacle avoidance method for UAVs for close-range autonomous exploration of the surface based on preset routes |
| CN109508016A (en)* | 2018-12-26 | 2019-03-22 | 北京工商大学 | Water quality sampling cruise ship path planning optimal method |
| CN109521794A (en)* | 2018-12-07 | 2019-03-26 | 南京航空航天大学 | A kind of multiple no-manned plane routeing and dynamic obstacle avoidance method |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016045615A1 (en)* | 2014-09-25 | 2016-03-31 | 科沃斯机器人有限公司 | Robot static path planning method |
| CN105974917A (en)* | 2016-05-11 | 2016-09-28 | 江苏大学 | Vehicle obstacle-avoidance path planning research method based on novel manual potential field method |
| CN106843235A (en)* | 2017-03-31 | 2017-06-13 | 深圳市靖洲科技有限公司 | It is a kind of towards the Artificial Potential Field path planning without person bicycle |
| WO2018176594A1 (en)* | 2017-03-31 | 2018-10-04 | 深圳市靖洲科技有限公司 | Artificial potential field path planning method for unmanned bicycle |
| CN108469828A (en)* | 2018-03-23 | 2018-08-31 | 哈尔滨工程大学 | A kind of AUV Route planners improving artificial potential field optimization algorithm |
| CN109358637A (en)* | 2018-05-25 | 2019-02-19 | 武汉科技大学 | A three-dimensional obstacle avoidance method for UAVs for close-range autonomous exploration of the surface based on preset routes |
| CN109521794A (en)* | 2018-12-07 | 2019-03-26 | 南京航空航天大学 | A kind of multiple no-manned plane routeing and dynamic obstacle avoidance method |
| CN109508016A (en)* | 2018-12-26 | 2019-03-22 | 北京工商大学 | Water quality sampling cruise ship path planning optimal method |
| Title |
|---|
| 刘琨等: "基于改进人工势场法的无人船路径规划算法", 《海南大学学报(自然科学版)》* |
| 刘琨等: "基于改进人工势场法的无人船路径规划算法", 《海南大学学报(自然科学版)》, no. 02, 30 May 2016 (2016-05-30)* |
| 师五喜等: "基于递阶模糊系统的人工势场法机器人路径规划", 《天津工业大学学报》* |
| 师五喜等: "基于递阶模糊系统的人工势场法机器人路径规划", 《天津工业大学学报》, no. 06, 25 December 2014 (2014-12-25)* |
| 陈金鑫等: "改进人工势场法的移动机器人路径规划", 《指挥控制与仿真》* |
| 陈金鑫等: "改进人工势场法的移动机器人路径规划", 《指挥控制与仿真》, no. 03, 15 January 2019 (2019-01-15)* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111123976A (en)* | 2019-12-24 | 2020-05-08 | 一飞智控(天津)科技有限公司 | Unmanned aerial vehicle cluster path planning processing method based on artificial potential field and unmanned aerial vehicle |
| US11545041B2 (en)* | 2020-02-28 | 2023-01-03 | Pablo Air Co., Ltd. | Method of avoiding collision of unmanned aerial vehicle |
| US20210272466A1 (en)* | 2020-02-28 | 2021-09-02 | Pablo Air Co., Ltd. | Method of avoiding collision of unmanned aerial vehicle |
| EP3872791B1 (en)* | 2020-02-28 | 2025-09-03 | Pablo Air Co., Ltd. | Method of avoiding collision of unmanned aerial vehicle |
| CN111506083A (en)* | 2020-05-19 | 2020-08-07 | 上海应用技术大学 | Safety obstacle avoidance method for industrial robot based on artificial potential field method |
| CN112150634A (en)* | 2020-08-31 | 2020-12-29 | 浙江工业大学 | A large-scale virtual scene roaming method based on multi-person redirection |
| CN112150634B (en)* | 2020-08-31 | 2024-03-26 | 浙江工业大学 | Large-scale virtual scene roaming method based on multi-person redirection |
| CN112923944A (en)* | 2021-01-29 | 2021-06-08 | 的卢技术有限公司 | Automatic driving path planning method and system and computer readable storage medium |
| CN113002537A (en)* | 2021-03-16 | 2021-06-22 | 镇江康飞汽车制造股份有限公司 | Vehicle active collision avoidance method based on artificial potential field method |
| CN113110441A (en)* | 2021-04-09 | 2021-07-13 | 江苏大学 | Agricultural unmanned vehicle cluster operation method based on ultra wide band |
| CN114077255A (en)* | 2021-11-22 | 2022-02-22 | 江苏理工学院 | A Way-finding Method for Intelligent Vehicles Based on Ellipse Model Artificial Potential Field Method |
| CN114265410A (en)* | 2021-12-25 | 2022-04-01 | 长安大学 | Local path planning method and system based on multi-computational power fusion |
| CN114460965A (en)* | 2022-01-21 | 2022-05-10 | 上海应用技术大学 | Unmanned aerial vehicle three-dimensional obstacle avoidance method based on improved artificial potential field method |
| CN114460965B (en)* | 2022-01-21 | 2023-08-29 | 上海应用技术大学 | Unmanned aerial vehicle three-dimensional obstacle avoidance method based on improved artificial potential field method |
| CN114442637B (en)* | 2022-02-10 | 2023-11-10 | 北京理工大学 | Unmanned vehicle local dynamic obstacle avoidance path planning method |
| CN114442637A (en)* | 2022-02-10 | 2022-05-06 | 北京理工大学 | A local dynamic obstacle avoidance path planning method for unmanned vehicles |
| CN115092183A (en)* | 2022-07-15 | 2022-09-23 | 东风柳州汽车有限公司 | Vehicle active obstacle avoidance control method and system based on potential field force |
| CN115092183B (en)* | 2022-07-15 | 2024-04-09 | 东风柳州汽车有限公司 | Active obstacle avoidance control method and system for vehicle based on potential field force |
| CN115416016A (en)* | 2022-08-15 | 2022-12-02 | 浙江工业大学 | Mechanical arm obstacle avoidance path planning method based on improved artificial potential field method |
| CN115793648A (en)* | 2022-11-29 | 2023-03-14 | 无锡学院 | An AGV path planning method |
| CN117369482A (en)* | 2023-12-06 | 2024-01-09 | 华润数字科技有限公司 | Path planning method, device and equipment for mobile robot and storage medium |
| CN117369482B (en)* | 2023-12-06 | 2024-03-12 | 华润数字科技有限公司 | Path planning method, device and equipment for mobile robot and storage medium |
| Publication number | Publication date |
|---|---|
| CN110567478B (en) | 2023-06-30 |
| Publication | Publication Date | Title |
|---|---|---|
| CN110567478B (en) | A Path Planning Method for Unmanned Vehicles Based on Artificial Potential Field Method | |
| CN112577491A (en) | Robot path planning method based on improved artificial potential field method | |
| CN106843235B (en) | A kind of Artificial Potential Field path planning towards no person bicycle | |
| CN109976340B (en) | Man-machine cooperation dynamic obstacle avoidance method and system based on deep reinforcement learning | |
| CN109501799A (en) | A kind of dynamic path planning method under the conditions of car networking | |
| CN111898211A (en) | Intelligent vehicle speed decision method and simulation method based on deep reinforcement learning | |
| JP4670807B2 (en) | Travel route creation method, autonomous mobile body, and autonomous mobile body control system | |
| CN106610664A (en) | Movement obstacle avoidance device and control method | |
| CN115469663B (en) | End-to-end navigation obstacle avoidance method based on deep reinforcement learning and oriented to automatic driving | |
| CN114594756A (en) | Multi-vehicle cooperative formation control method, terminal and medium in dynamic obstacle environment | |
| CN113190004A (en) | Unmanned vehicle collision avoidance path planning method based on algorithm fusion | |
| Zhang et al. | Structured road-oriented motion planning and tracking framework for active collision avoidance of autonomous vehicles | |
| Yang et al. | Path planning and trajectory optimization based on improved APF and multi-target | |
| CN112327831A (en) | Factory AGV track planning method based on improved artificial potential field method | |
| Zhuge et al. | A novel dynamic obstacle avoidance algorithm based on collision time histogram | |
| CN115077553A (en) | Grid-based search trajectory planning method, system, vehicle, equipment and medium | |
| CN114442628A (en) | Path planning method, device and system for mobile robot based on artificial potential field method | |
| CN118730145A (en) | A path planning method for park logistics vehicles based on map-free navigation | |
| Liu et al. | Research on local real-time obstacle avoidance path planning of unmanned vehicle based on improved artificial potential field method | |
| CN119354217A (en) | A path planning method based on improved artificial potential field method | |
| CN117685978A (en) | Self-adaptive obstacle avoidance path planning method and system | |
| CN115092183B (en) | Active obstacle avoidance control method and system for vehicle based on potential field force | |
| CN117215307A (en) | Unmanned vehicle track planning method based on improved field theory | |
| CN116572960A (en) | Bypass control method, device, terminal equipment and storage medium | |
| CN117864135A (en) | Vehicle lane-changing steering assistance method based on driver's operating comfort |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right | Effective date of registration:20240313 Address after:211800, No. 9-156, Buyue Road, Pukou Economic Development Zone, Pukou District, Nanjing City, Jiangsu Province Patentee after:Qingyan Intelligent Technology (Nanjing) Co.,Ltd. Country or region after:China Address before:545006 Guangxi University of science and technology, 268 Donghuan Avenue, Liuzhou City, Guangxi Zhuang Autonomous Region Patentee before:GUANGXI University OF SCIENCE AND TECHNOLOGY Country or region before:China |