技术领域technical field
本发明涉及人工智能算法技术领域，更具体的涉及一种基于碰撞风险分析的四轮移动机器人模糊避障方法。The invention relates to the technical field of artificial intelligence algorithms, and in particular to a fuzzy obstacle avoidance method for a four-wheeled mobile robot based on collision risk analysis.
背景技术Background technique
前轮导向、后轮驱动的四轮移动机器人(Front-wheel-Oriented&Rear-wheel-Driven FWMR,下文简称为FORD-FWMR)是货物搬运机器人、温室移动机器人和农药喷洒机器人等机器人应用的首选方案，对于智能仓储和智能农业等领域的未来发展具有重要意义。为了在提高工作效率和节能减排等方面更好地发挥作用，FORD-FWMR的自主避障逐渐成为国内外学者的研究热点之一。Front-wheel-Oriented&Rear-wheel-Driven FWMR (hereinafter referred to as FORD-FWMR) is the preferred solution for robot applications such as cargo handling robots, greenhouse mobile robots and pesticide spraying robots. It is of great significance to the future development of fields such as intelligent storage and intelligent agriculture. In order to play a better role in improving work efficiency and energy saving and emission reduction, the autonomous obstacle avoidance of FORD-FWMR has gradually become one of the research hotspots of scholars at home and abroad.
在机器人运行环境未知时，模糊逻辑法由于不需要建立运行环境的精确数学模型，算法实时性较好且不存在局部极小问题，因此被广泛应用于移动机器人的避障研究。然而，传统模糊避障算法的输入项往往只包括障碍物在不同方向的距离信息。例如，江贵龙以左侧和右侧障碍物的距离为输入，提出了一种可有效降低障碍物幻影干扰的移动机器人模糊避障算法，申永红以左、右、前三个方向上障碍物的距离为输入，构建了一种用于智能小车自主避障的模糊控制器；丁吉以左前方、右前方障碍物的距离和机器人与障碍物的位置关系为输入，设计了一种针对三轮移动机器人在未知环境下避障问题的模糊规划器，但上述三者均没有将速度因素作为模糊避障决策过程的输入，在障碍物密集环境下仍然可能发生碰撞或选择不合理路径，且江贵龙的算法控制规则过于粗糙，容易陷入死循环。大多数直接引入速度反馈输入项的模糊避障算法则增大了输入数量和模糊规则数量，进而增大了算法复杂度和存储空间需求，也提高了建立合理模糊规则的难度和引入人类主观误差的可能性；如果为降低模糊规则而刻意减少各输入项的模糊语言集合元素数量，又将显著影响避障算法的精度和效果，陷入两难境地。例如，彭玉青针对履带式移动机器人提出了一种包含速度反馈的模糊避障算法，提高了障碍物密集环境下的避障性能，但需要结合红外信标、红外接收器等额外硬件来实现防迂回功能。此外，该算法为控制规则数量，将3个模糊输入项(前、左、右方向上的障碍物距离)的模糊语言集合元素数量同时精简为2个({近，远})，粗糙的距离区分度很可能影响最终的模糊控制精度，与此同时整个模糊控制规则仍然达到了24个(速度反馈输入的模糊语言集合为{慢、中、快})。When the operating environment of the robot is unknown, the fuzzy logic method is widely used in the research of obstacle avoidance of mobile robots because it does not need to establish an accurate mathematical model of the operating environment, the algorithm has good real-time performance and there is no local minimum problem. However, the input items of traditional fuzzy obstacle avoidance algorithms often only include the distance information of obstacles in different directions. For example, Jiang Guilong used the distance of obstacles on the left and right as input, and proposed a fuzzy obstacle avoidance algorithm for mobile robots that can effectively reduce the interference of phantom obstacles. Shen Yonghong used the distances of obstacles in the left, right, and front directions As input, a fuzzy controller for autonomous obstacle avoidance of smart cars was constructed; Ding Ji took the distance of obstacles in the left and right fronts and the positional relationship between the robot and obstacles as inputs, and designed a three-wheel mobile Fuzzy planners for robot obstacle avoidance in unknown environments, but none of the above three uses the speed factor as the input of fuzzy obstacle avoidance decision-making process, collisions or unreasonable paths may still occur in dense obstacle environments, and Jiang Guilong's Algorithmic control rules are too rough and easy to fall into an endless loop. Most fuzzy obstacle avoidance algorithms that directly introduce speed feedback input items increase the number of inputs and fuzzy rules, thereby increasing the complexity of the algorithm and storage space requirements, and also increasing the difficulty of establishing reasonable fuzzy rules and introducing human subjective errors. possibility; if the number of fuzzy language set elements of each input item is deliberately reduced in order to reduce the fuzzy rules, it will significantly affect the accuracy and effect of the obstacle avoidance algorithm and fall into a dilemma. For example, Peng Yuqing proposed a fuzzy obstacle avoidance algorithm including speed feedback for tracked mobile robots, which improves the obstacle avoidance performance in dense obstacle environments, but requires additional hardware such as infrared beacons and infrared receivers to achieve anti-obstacle. detour function. In addition, in order to control the number of rules, the algorithm reduces the number of fuzzy language set elements of 3 fuzzy input items (obstacle distances in the front, left, and right directions) to 2 ({near, far}) at the same time, and the rough distance The degree of discrimination is likely to affect the final fuzzy control accuracy, and at the same time the whole fuzzy control rules still reach 24 (the fuzzy language set of speed feedback input is {slow, medium, fast}).
综上所述，现有技术的主要缺陷：缺少一种针对前轮导向、后轮驱动的四轮移动机器人在障碍物密集环境中的高效模糊避障算法，能够同时较好地满足低规则复杂度和计算成本，以及高避障决策与控制性能两个条件。In summary, the main defect of the existing technology is the lack of an efficient fuzzy obstacle avoidance algorithm for front-wheel-guided, rear-wheel-driven four-wheel mobile robots in dense obstacle environments, which can better meet the requirements of low-rule and complex algorithms at the same time. Accuracy and computational cost, as well as high obstacle avoidance decision-making and control performance are two conditions.
发明内容Contents of the invention
本发明实施例提供一种基于碰撞风险分析的四轮移动机器人模糊避障方法，用以解决上述背景技术中存在的问题。An embodiment of the present invention provides a fuzzy obstacle avoidance method for a four-wheeled mobile robot based on collision risk analysis to solve the problems in the background technology above.
本发明实施例提供一种基于碰撞风险分析的四轮移动机器人模糊避障方法，包括：An embodiment of the present invention provides a fuzzy obstacle avoidance method for a four-wheeled mobile robot based on collision risk analysis, including:
获取要素：四轮移动机器人与障碍物之间的距离、四轮移动机器人与障碍物的相对运动角度、以及四轮移动机器人与障碍物的相对速度；Obtaining elements: the distance between the four-wheel mobile robot and the obstacle, the relative motion angle between the four-wheel mobile robot and the obstacle, and the relative speed between the four-wheel mobile robot and the obstacle;
通过多要素碰撞风险模型，确定当前四轮移动机器人左前方、右前方的碰撞风险LRC和RRC；Through the multi-element collision risk model, determine the collision risks LRC and RRC of the left front and right front of the current four-wheel mobile robot;
根据模糊控制规则，将当前四轮移动机器人左前方、右前方的碰撞风险LRC和RRC输入至模糊避障控制器，获得由模糊避障控制器输出的下一时刻四轮移动机器人的前轮转角Φf和后轮转速vb；According to the fuzzy control rules, input the collision risks LRC and RRC of the left front and right front of the current four-wheel mobile robot to the fuzzy obstacle avoidance controller, and obtain the front of the four-wheel mobile robot at the next moment output by the fuzzy obstacle avoidance controller. Wheel rotation angle Φf and rear wheel speed vb ;
对下一时刻四轮移动机器人的前轮转角Φf和后轮转速vb进行去模糊化处理，并将模糊化处理后的下一时刻四轮移动机器人的前轮转角Φf和后轮转速vb引入四轮移动机器人的运动学模型中，确定实际四轮移动机器人的左、右前轮转角和左、右后轮转速；Defuzzify the front wheel angle Φf and the rear wheel speed v b of the four-wheel mobile robot at the next moment, and defuzzify the front wheel angle Φf and rear wheel speed vb of the four-wheel mobile robot at the next moment vb is introduced into the kinematics model of the four-wheel mobile robot to determine the rotation angle of the left and right front wheels and the rotational speed of the left and right rear wheels of the actual four-wheel mobile robot;
将实际四轮移动机器人的左、右前轮转角和左、右后轮转速输入至四轮移动机器人的驱动机构，控制四轮移动机器人进行模糊避障。The left and right front wheel rotation angles and the left and right rear wheel speeds of the actual four-wheel mobile robot are input to the driving mechanism of the four-wheel mobile robot to control the four-wheel mobile robot for fuzzy obstacle avoidance.
进一步地，所述多要素碰撞风险模型，具体包括：Further, the multi-element collision risk model specifically includes:
假设四轮移动机器人通过自身多个传感器能够实时获取正前方180°范围内障碍物的距离和方位信息，对四轮移动机器人左前方和右前方一定角度范围θmax内的障碍物进行碰撞风险分析，并对四轮移动机器人左、右两侧θmax角度范围之外障碍物的碰撞风险用距离表示；其中，θmax取值范围为[45°,70°]；则四轮移动机器人与障碍物或密集障碍物区域的多要素碰撞风险RC(i)的计算模型如下所示：Assuming that the four-wheeled mobile robot can obtain the distance and orientation information of obstacles within a range of 180° in real time through its own multiple sensors, the collision risk analysis of obstacles within a certain angle range θmax in the left and right fronts of the four-wheeled mobile robot is carried out. , and the collision risk of obstacles outside the angle range of θmax on the left and right sides of the four-wheel mobile robot is represented by distance; where, the value range of θmax is [45°,70°]; then the four-wheel mobile robot and the obstacle The calculation model of the multi-element collision risk RC(i) in the area of obstacles or dense obstacles is as follows:
RC(i)＝aD(i)+bF(i)D(i)+cV(i)F(i)D(i)RC(i)=aD(i)+bF(i)D(i)+cV(i)F(i)D(i)
其中，a、b、c为D(i)、F(i)和V(i)的影响因子，且a＝b＝c＝1；D(i)反映RC(i)与障碍物距离di的关系；F(i)反映RC(i)与障碍物相对运动角度θi的关系；V(i)反映RC(i)与障碍物相对速度vi的关系；dmax为最大距离，dmin为安全距离，di＝min(dif,dir,dil)；θmax为碰撞风险分析的最大角度,θmin为安全角度，θmin＝max(θAOC,θBOC)；vmax为安全速度,vmin为最小速度；di为四轮移动机器人与障碍物之间的最短距离，即前、右、左方向上距离测量结果的最小值；为向量，即四轮移动机器人与障碍物的相对速度；A、B为障碍物相对于四轮移动机器人上传感器能感知的两个最边缘点，与障碍物的空间尺寸相关；C为障碍物与四轮移动机器人距离最近的点；θi为的方向与直线OC之间的夹角，即四轮移动机器人与障碍物的相对运动角度；当θi一定时，θmin由障碍物相对于超声波传感器能感知的两个最边缘点A和B所确定，而夹角θAOC和θBOC与障碍物或密集障碍物区域的空间尺寸有关。Among them, a, b, c are the influencing factors of D(i), F(i) and V(i), and a=b=c=1; D(i) reflects the distance di between RC(i) and obstacles F(i) reflects the relationship between RC(i) and the relative movement angle θi of the obstacle; V(i) reflects the relationship between RC(i) and the relative velocity vi of the obstacle; dmax is the maximum distance, dmin is the safety distance, di =min(dif ,dir ,dil ); θmax is the maximum angle of collision risk analysis, θmin is the safety angle, θmin =max(θAOC ,θBOC ); vmax is Safe speed, vmin is the minimum speed; di is the shortest distance between the four-wheel mobile robot and the obstacle, that is, the minimum value of the distance measurement results in the front, right and left directions; is the vector, that is, the relative speed between the four-wheeled mobile robot and the obstacle; A and B are the two most edge points of the obstacle relative to the sensors on the four-wheeled mobile robot, which are related to the spatial size of the obstacle; C is the obstacle The point closest to the four-wheeled mobile robot; θi is The angle between the direction and the straight line OC, that is, the relative motion angle between the four-wheeled mobile robot and the obstacle; when θi is constant, θmin is determined by the two most edge points A and B that the obstacle can perceive relative to the ultrasonic sensor determined, and the included angles θAOC and θBOC are related to the spatial size of obstacles or dense obstacle areas.
进一步地，采用三角函数作为当前四轮移动机器人左前方、右前方的碰撞风险LRC和RRC的隶属度函数；以及采用三角函数作为下一时刻四轮移动机器人的前轮转角Φf和后轮转速vb的隶属度函数。Further, trigonometric functions are used as the membership function of the collision risks LRC and RRC of the current four-wheeled mobile robot’s left and right fronts; and trigonometric functions are used as the front wheel angle Φf and the rear The membership function of the wheel speed vb .
进一步地，所述模糊控制规则，具体包括：Further, the fuzzy control rules specifically include:
当四轮移动机器人远离距离障碍物时，保持原方向继续高速行驶；When the four-wheel mobile robot is far away from the obstacle, keep the original direction and continue to drive at high speed;
当四轮移动机器人左前方靠近障碍物时，减速并向右转向，反之则减速并向左转向，转向程度大小由左前方或右前方障碍物的距离、相对角度和相对速度决定；When the front left of the four-wheel mobile robot is close to an obstacle, it will slow down and turn to the right, otherwise it will slow down and turn to the left. The degree of steering is determined by the distance, relative angle and relative speed of the obstacle in front left or right;
当障碍物位于四轮移动机器人正前方或左、右碰撞风险基本相同时，优先向右转向。When the obstacle is located directly in front of the four-wheel mobile robot or the left and right collision risks are basically the same, turn to the right first.
进一步地，所述对下一时刻四轮移动机器人的前轮转角Φf和后轮转速vb进行去模糊化处理；具体包括：Further, the defuzzification process is performed on the front wheel angle Φf and the rear wheel speed vb of the four-wheel mobile robot at the next moment; specifically includes:
采用面积重心法进行去模糊化处理，具体过程如下：The area center of gravity method is used for defuzzification, and the specific process is as follows:
上式中的∫表示输出模糊子集所有元素的隶属度值在连续论域上的代数积分，μΦ、μv分别表示前轮转角Φf和后轮转速vb的隶属度函数。∫ in the above formula represents the algebraic integral of the membership degree values of all elements of the output fuzzy subset on the continuous universe, and μΦ and μv represent the membership function of the front wheel rotation angle Φf and the rear wheel speed vb respectively.
进一步地，所述四轮移动机器人的运动学模型，具体包括：Further, the kinematics model of the four-wheeled mobile robot specifically includes:
假设前轮转角用Φf表示，以四轮移动机器人的前进方向作为参考，设定左转时Φf为正，右转时Φf为负，则左前轮转角Φl和Φf的关系：Assuming that the front wheel rotation angle is represented by Φf , and taking the forward direction of the four-wheel mobile robot as a reference, Φf is set to be positive when turning left, and Φ fis negative when turning right, then the relationship between the left front wheel rotation angle Φl and Φf :
式中，L为前、后轮轴距，w为四轮移动机器人宽度；In the formula, L is the wheelbase of the front and rear wheels, and w is the width of the four-wheel mobile robot;
Φl与两后轮中点处转弯半径R0的关系为：The relationship between Φl and the turning radius R0 at the midpoint of the two rear wheels is:
R0＝L cotΦl+w/2R0 ＝L cotΦl +w/2
右前轮转角Φr与两后轮中点处转弯半径R0的关系为：The relationship between the turning angle Φr of the right front wheel and the turning radius R0 at the midpoint of the two rear wheels is:
R0＝L cotΦr-w/2R0 ＝L cotΦr -w/2
前轮左转时左、右两后轮的转速比m为：When the front wheels turn left, the speed ratio m of the left and right rear wheels is:
式中，Rl为左后轮转弯半径，Rr为右后轮转弯半径，cl为左后轮与机器人基体距离，cr为右后轮与机器人基体距离；In the formula, Rl is the turning radius of the left rear wheel, Rr is the turning radius of the right rear wheel, cl is the distance between the left rear wheel and the robot base, and cr is the distance between the right rear wheel and the robot base;
前轮右转时左、右两后轮的转速比m为When the front wheels turn right, the speed ratio m of the left and right rear wheels is
设四轮移动机器人的后轮转速为vb，则左、右后轮的转速nl和nr分别为：Assuming that the speed of the rear wheel of the four-wheel mobile robot is vb , the speeds nl and nr of the left and right rear wheels are respectively:
式中，r为四轮移动机器人的车轮半径。In the formula, r is the wheel radius of the four-wheeled mobile robot.
本发明实施例提供一种基于碰撞风险分析的四轮移动机器人模糊避障方法，与现有技术相比，其有益效果如下：An embodiment of the present invention provides a fuzzy obstacle avoidance method for a four-wheeled mobile robot based on collision risk analysis. Compared with the prior art, the beneficial effects are as follows:
本发明针对前轮导向、后轮驱动的四轮移动机器人在障碍物密集环境下的模糊避障需要同时考虑障碍物的距离、方向、尺寸和相对速度等多种因素，导致模糊控制器输入项和控制规则数量过大的问题，提出一种基于碰撞风险分析的双输入模糊避障算法。通过建立包含障碍物的距离、尺寸、相对运动角度和相对速度等因素的碰撞风险模型，以当前时刻机器人左前方和右前方的碰撞风险估计结果为输入，以下一时刻前轮转角和后轮转速为输出，结合建立的运动学模型计算左、右前轮转角和左、右后轮转速，输入驱动机构以实现避障动作。该算法能够在降低生成的模糊控制器的控制规则数量和运行成本的前提下，实现较高精度的自主避障，即相对于同类型算法，本发明能够在降低模糊避障控制算法的规则数量和计算成本的同时，提高整个避障决策过程中障碍物分析的全面性和精细程度，进一步改善控制性能。The present invention aims at the fuzzy obstacle avoidance of a four-wheeled mobile robot with front-wheel guidance and rear-wheel drive in a dense environment of obstacles, which needs to consider multiple factors such as the distance, direction, size and relative speed of obstacles at the same time, resulting in fuzzy controller input items Aiming at the problem of excessive number of control rules and control rules, a dual-input fuzzy obstacle avoidance algorithm based on collision risk analysis is proposed. By establishing a collision risk model that includes factors such as obstacle distance, size, relative motion angle, and relative speed, and taking the collision risk estimation results of the left and right front of the robot at the current moment as input, the front wheel rotation angle and rear wheel speed at the next moment As the output, the left and right front wheel rotation angles and the left and right rear wheel speeds are calculated in combination with the established kinematics model, and input to the driving mechanism to realize the obstacle avoidance action. The algorithm can realize high-precision autonomous obstacle avoidance under the premise of reducing the number of control rules and operating costs of the generated fuzzy controller, that is, compared with algorithms of the same type, the present invention can reduce the number of rules of the fuzzy obstacle avoidance control algorithm While calculating the cost, the comprehensiveness and precision of obstacle analysis in the entire obstacle avoidance decision-making process are improved, and the control performance is further improved.
附图说明Description of drawings
图1为本发明实施例提供的碰撞分析范围划分图；FIG. 1 is a division diagram of a collision analysis range provided by an embodiment of the present invention;
图2为本发明实施例提供的FWMR与障碍物的相对位置示意图；Fig. 2 is a schematic diagram of the relative positions of FWMR and obstacles provided by the embodiment of the present invention;
图3为本发明实施例提供的模糊逻辑控制器结构图；Fig. 3 is the structural diagram of the fuzzy logic controller that the embodiment of the present invention provides;
图4(a)为本发明实施例提供的碰撞风险LRC和RRC的隶属度函数；Fig. 4 (a) is the membership function of collision risk LRC and RRC provided by the embodiment of the present invention;
图4(b)为本发明实施例提供的前轮转角Φf的隶属度函数；Fig. 4 (b) is the membership function of the front wheel rotation angle Φf provided by the embodiment of the present invention;
图4(c)为本发明实施例提供的后轮转速vb的隶属度函数。Fig. 4(c) is the membership function of the rear wheel speed vb provided by the embodiment of the present invention.
具体实施方式Detailed ways
下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
参见图1～4，本发明实施例提供一种基于碰撞风险分析的四轮移动机器人模糊避障方法，该方法包括：Referring to Figures 1-4, an embodiment of the present invention provides a fuzzy obstacle avoidance method for a four-wheeled mobile robot based on collision risk analysis, the method comprising:
步骤1：获取要素：四轮移动机器人与障碍物之间的距离、四轮移动机器人与障碍物的相对运动角度、以及四轮移动机器人与障碍物的相对速度。Step 1: Obtain elements: the distance between the four-wheel mobile robot and the obstacle, the relative motion angle between the four-wheel mobile robot and the obstacle, and the relative speed between the four-wheel mobile robot and the obstacle.
步骤2：通过多要素碰撞风险模型，确定当前四轮移动机器人左前方、右前方的碰撞风险LRC和RRC。Step 2: Through the multi-element collision risk model, determine the collision risks LRC and RRC of the left front and right front of the current four-wheel mobile robot.
步骤3：根据模糊控制规则，将当前四轮移动机器人左前方、右前方的碰撞风险LRC和RRC输入至模糊避障控制器，获得由模糊避障控制器输出的下一时刻四轮移动机器人的前轮转角Φf和后轮转速vb。Step 3: According to the fuzzy control rules, input the collision risks LRC and RRC of the left front and right front of the current four-wheel mobile robot to the fuzzy obstacle avoidance controller, and obtain the four-wheel movement at the next moment output by the fuzzy obstacle avoidance controller The robot's front wheel rotation angle Φf and rear wheel speed vb .
步骤4：对下一时刻四轮移动机器人的前轮转角Φf和后轮转速vb进行去模糊化处理，并将下一时刻四轮移动机器人的前轮转角Φf和后轮转速vb引入四轮移动机器人的运动学模型中，确定四轮移动机器人的左、右前轮转角和左、右后轮转速。Step 4: Defuzzify the front wheel angle Φf and the rear wheel speed vb of the four-wheel mobile robot at the next moment, and calculate the front wheel angle Φf and rear wheel speed vb of the four-wheel mobile robot at the next moment Introduce the kinematics model of the four-wheel mobile robot to determine the left and right front wheel rotation angles and the left and right rear wheel speeds of the four-wheel mobile robot.
步骤5：将四轮移动机器人的左、右前轮转角和左、右后轮转速输入至四轮移动机器人的驱动机构，控制四轮移动机器人进行模糊避障。Step 5: Input the left and right front wheel rotation angles and the left and right rear wheel speeds of the four-wheel mobile robot to the drive mechanism of the four-wheel mobile robot to control the four-wheel mobile robot for fuzzy obstacle avoidance.
上述步骤1～5，具体过程如下：The above steps 1 to 5, the specific process is as follows:
1、建立前轮导向、后轮驱动四轮移动机器人的运动学模型1. Establish the kinematics model of the four-wheel mobile robot with front-wheel guidance and rear-wheel drive
假设前轮转角用Φf表示，以FORD-FWMR的前进方向作为参考，设定左转时Φf为正，右转时Φf为负，则左前轮转角Φl和Φf的关系为：Assuming that the front wheel rotation angle is represented by Φf , and taking the forward direction of FORD-FWMR as a reference, Φf is set to be positive when turning left, and Φ fis negative when turning right, then the relationship between the left front wheel rotation angle Φl and Φf is :
式(1)中，L为前、后轮轴距，w为机器人宽度。In formula (1), L is the wheelbase of the front and rear wheels, and w is the width of the robot.
Φl与两后轮中点处转弯半径R0的关系为The relationship between Φl and the turning radius R0 at the midpoint of the two rear wheels is
R0＝L cotΦl+w/2 (2)R0 ＝L cotΦl +w/2 (2)
右前轮转角Φr与两后轮中点处转弯半径R0的关系为The relationship between the turning angle Φr of the right front wheel and the turning radius R0 at the midpoint of the two rear wheels is
R0＝L cotΦr-w/2 (3)R0 ＝L cotΦr -w/2 (3)
前轮左转时左、右两后轮的转速比m为When the front wheels turn left, the rotational speed ratio m of the left and right rear wheels is
式(4)中，Rl为左后轮转弯半径，Rr为右后轮转弯半径，cl为左后轮与机器人基体距离，cr为右后轮与机器人基体距离。In formula (4), Rl is the turning radius of the left rear wheel, Rr is the turning radius of the right rear wheel, cl is the distance between the left rear wheel and the robot base, andcr is the distance between the right rear wheel and the robot base.
同理，前轮右转时左、右两后轮的转速比m为Similarly, when the front wheels turn right, the rotational speed ratio m of the left and right rear wheels is
设FORD-FWMR后轮转速为vb，左、右后轮的转速nl和nr分别为Assuming that the rear wheel speed of FORD-FWMR is vb , the speeds nl and nr of the left and right rear wheels are respectively
式(6)和式(7)中，r为FORD-FWMR的车轮半径。In formula (6) and formula (7), r is the wheel radius of FORD-FWMR.
2、建立碰撞风险分析模型2. Establish a collision risk analysis model
算法利用碰撞风险(Risk of Collision,RC)描述移动机器人与障碍物之间发生碰撞的危险程度并作为模糊控制器的输入。移动机器人与障碍物的碰撞风险与障碍物的距离、尺寸、相对运动角度和相对速度有关。一般来说，障碍物的距离越小、尺寸越大、相对运动角度越小、相对速度越大，则碰撞风险越大；反之则越小。假设移动机器人通过自身多个传感器能够实时获取正前方180°范围内障碍物详细的距离和方位信息，避障动作主要针对前方一定角度范围(<180°)内的障碍物，无需对该角度范围之外的障碍物进行精确描述。因此，对移动机器人前方180°范围进行如图1所示的划分，对左前方和右前方一定角度范围(θmax)内的障碍物进行精确的碰撞风险分析，左、右两侧θmax角度范围之外障碍物的碰撞风险则直接用距离表示。θmax的参考取值范围为[45°,70°]。The algorithm uses the risk of collision (Risk of Collision, RC) to describe the risk of collision between the mobile robot and the obstacle, and it is used as the input of the fuzzy controller. The collision risk between a mobile robot and an obstacle is related to the distance, size, relative motion angle and relative speed of the obstacle. Generally speaking, the smaller the distance, the larger the size, the smaller the relative motion angle, and the larger the relative speed of the obstacle, the greater the risk of collision; vice versa. Assuming that the mobile robot can obtain detailed distance and orientation information of obstacles within 180° in real time through its own multiple sensors, the obstacle avoidance action is mainly aimed at obstacles within a certain angle range (<180°) in front, and there is no need for this angle range Accurate description of obstacles outside. Therefore, the 180° range in front of the mobile robot is divided as shown in Figure 1, and an accurate collision risk analysis is performed on obstacles within a certain angle range (θmax ) in the left and rightfronts . The collision risk of obstacles outside the range is directly expressed in distance. The reference value range of θmax is [45°, 70°].
假设移动机器人(FWMR)与障碍物或密集障碍物区域(Obsti)的相对位置如图2所示。图2中各符号的具体含义如下：Assume that the relative position of the mobile robot (FWMR) and the obstacle or dense obstacle area (Obsti ) is shown in Figure 2. The specific meanings of the symbols in Figure 2 are as follows:
di—FWMR与Obsti之间的最短距离，即前、右、左方向上距离测量结果的最小值；—向量，FWMR与Obsti的相对速度；A,B—Obsti相对于FWMR上传感器可感知的两个最边缘点，与障碍物的空间尺寸相关；C—Obsti上与FWMR距离最近的点；θi—的方向与直线OC之间的夹角，FWMR与Obsti的相对运动角度。则碰撞风险RC(i)的计算模型如下：di —the shortest distance between FWMR and Obsti , that is, the minimum value of the distance measurement results in the front, right and left directions; —Vector, the relative speed between FWMR and Obsti ; A, B—the two most edge points of Obsti relative to the sensor on FWMR, which are related to the spatial size of the obstacle; C—the closest point on Obsti to FWMR ; θi — The angle between the direction of and the straight line OC, the relative movement angle between FWMR and Obsti . Then the calculation model of collision risk RC(i) is as follows:
RC(i)＝aD(i)+bF(i)D(i)+cV(i)F(i)D(i) (8)RC(i)=aD(i)+bF(i)D(i)+cV(i)F(i)D(i) (8)
其中，a、b、c为D(i)、F(i)和V(i)的影响因子，建议a＝b＝c＝1；D(i)反映RC(i)与障碍物距离di的关系，F(i)反映RC(i)与障碍物相对运动角度θi的关系，V(i)反映RC(i)与障碍物相对速度vi的关系。dmax为最大距离，dmin为安全距离，di＝min(dif,dir,dil)；θmax为图2中的精确碰撞风险分析的最大角度,θmin为安全角度，θmin＝max(θAOC,θBOC)；vmax为安全速度,vmin为最小速度。Among them, a, b, c are the influencing factors of D(i), F(i) and V(i), it is suggested that a=b=c=1; D(i) reflects the distance di between RC(i) and obstacles F(i) reflects the relationship between RC(i) and the relative motion angle θi of the obstacle, and V(i) reflects the relationship between RC(i) and the relative velocity vi of the obstacle. dmax is the maximum distance, dmin is the safety distance, di =min(dif ,dir ,dil ); θmax is the maximum angle of accurate collision risk analysis in Figure 2, θmin is the safety angle, θmin =max(θAOC ,θBOC ); vmax is the safe speed, vmin is the minimum speed.
需要特别指出的是，当θi一定时，θmin由障碍物相对于超声波传感器可感知的两个最边缘点A和B所确定，而夹角θAOC和θBOC和障碍物或密集障碍物区域的空间尺寸有关，因此式(10)实际上同时包含了障碍物或密集障碍物区域的相对运动角度信息和空间尺寸信息。It should be pointed out that when θi is constant, θmin is determined by the two most edge points A and B of the obstacle relative to the ultrasonic sensor, and the angle θAOC and θBOC and the obstacle or dense obstacle The spatial size of the area is related, so the formula (10) actually contains the relative motion angle information and the spatial size information of the obstacle or dense obstacle area at the same time.
3、构建模糊控制器3. Build a fuzzy controller
3.1输入输出及其模糊语言描述3.1 Input and output and its fuzzy language description
首先确定模糊控制器的输入。本文算法基于碰撞风险计算式(8-11)，利用前、右、左方向上的障碍物距离、方位信息与速度反馈信息计算当前时刻机器人左前方和右前方的碰撞风险LRC和RRC，并将其作为作为模糊避障控制器的输入。First determine the input of the fuzzy controller. The algorithm in this paper is based on the collision risk calculation formula (8-11), using the obstacle distance, orientation information and speed feedback information in the front, right, and left directions to calculate the collision risks LRC and RRC of the left and right front of the robot at the current moment, And take it as the input of the fuzzy obstacle avoidance controller.
其次确定模糊控制器的输出。对于FORD-FWMR来说，只要由模糊控制器输出前轮转角Φf和后轮转速vb，即可基于式(1)、(3)分别确定左、右前轮的转角Φl和Φr，基于式(6)、(7)分别确定左、右后轮的转速nl和nr，实现FORD-FWMR在避障导航过程中转向与变速的协同控制。因此，选择Φf和vb作为模糊避障控制器的输出。Second, determine the output of the fuzzy controller. For FORD-FWMR, as long as the fuzzy controller outputs the front wheel rotation angle Φf and the rear wheel speed vb , the left and right front wheel rotation angles Φl and Φr can be determined based on equations (1) and (3) respectively , based on equations (6) and (7), respectively determine the rotational speeds nl and nr of the left and right rear wheels, and realize the coordinated control of steering and gear shifting in FORD-FWMR during obstacle avoidance navigation. Therefore, choose Φf and vb as the output of the fuzzy obstacle avoidance controller.
下面确定输入输出量的模糊语言描述。根据公式(8)可知，若取a＝1，b＝1，c＝1，则LRC和RRC的论域均为[0,3]，模糊子集均确定为{Z,S,M,B}，分别表示碰撞危险度为{无，小，中，大}。Φf的论域为[-π/2,π/2]，设左偏为负，右偏为正，则相应的模糊语言变量记作{NB，NM，NS，NVS，ZO，PVS，PS，PM，PB}，分别表示转向角度为{左大，左中，左小，左很小，正前，右很小，右小，右中，右大}；vb的论域为[0.1m/s，0.3m/s]，相应的模糊语言变量记作{LV，MV，HV}，分别代表运行速度为{低速，中速，高速}。综上所述，模糊避障控制的结构如图3所示。The fuzzy language description of the input and output quantities is determined below. According to the formula (8), if a=1, b=1, c=1, then the domains of LRC and RRC are both [0,3], and the fuzzy subsets are determined as {Z, S, M ,B}, which respectively represent the collision risk as {none, small, medium, large}. The domain of Φf is [-π/2,π/2], if the left bias is negative and the right bias is positive, then the corresponding fuzzy language variables are recorded as {NB, NM, NS, NVS, ZO, PVS, PS , PM, PB}, which respectively indicate that the steering angle is {large left, middle left, small left, small left, small right, small right, small right, middle right, large right}; the domain of vb is [0.1 m/s, 0.3m/s], and the corresponding fuzzy language variables are recorded as {LV, MV, HV}, respectively representing the running speed as {low speed, medium speed, high speed}. In summary, the structure of fuzzy obstacle avoidance control is shown in Figure 3.
3.2隶属度函数3.2 Membership function
为了减少模糊控制计算量，提高响应速度，选择三角函数作为隶属度函数。输入项LRC和RRC的隶属度函数均如图4(a)所示；输出项前轮转角Φf的隶属度函数如图4(b)所示；输出项后轮转速vb的隶属度函数如图4(c)所示。In order to reduce the calculation amount of fuzzy control and improve the response speed, the trigonometric function is selected as the membership function. The membership functions of the input items LRC and RRC are shown in Fig. 4(a); the membership function of the output item front wheel rotation angle Φf is shown in Fig. 4(b); the membership function of the output item rear wheel speed vb The degree function is shown in Fig. 4(c).
3.3模糊控制规则3.3 Fuzzy control rules
建立模糊控制规则是实现模糊控制的关键步骤。根据日常驾驶经验，距离障碍物很远时应保持原方向继续高速行驶；左前方距离障碍物较近时，应减速并向右转向，反之则应减速并向左转向，转向程度大小由左前方或右前方障碍物的距离、相对角度(方位)和相对速度等因素综合决定。特别地，如果障碍物位于机器人正前方或左、右碰撞风险基本相同时，则规定优先向右转向。根据输入项数量(LRC和RRC共2个输入项)和每个输入项的模糊语言变量数量(均为Z,S,M,B等4个模糊变量)，生成的模糊控制器共有4×4＝16条控制规则。Establishing fuzzy control rules is the key step to realize fuzzy control. According to daily driving experience, when you are far away from obstacles, you should keep the original direction and continue driving at high speed; when the left front is close to obstacles, you should slow down and turn to the right; otherwise, you should slow down and turn to the left. Or the distance of the obstacle in front of the right, the relative angle (azimuth) and relative speed and other factors are comprehensively determined. In particular, if the obstacle is located directly in front of the robot or the left and right collision risks are basically the same, it is stipulated to turn to the right first. According to the number of input items (LRC and RRC have 2 input items) and the number of fuzzy language variables of each input item (all are Z, S, M, B, etc. 4 fuzzy variables), the generated fuzzy controller has a total of 4 ×4=16 control rules.
控制规则的描述统一使用“IF-THEN”语句，其一般形式为IF(LRC is LRCi and RRCis RRCj)THEN(ΦfisΦfm and vb is vbn)。其中，LRCi、RRCj、Φfm和vbn分别为定义在论域LRC、RRC、Φf和vb上的模糊集。The description of the control rules uniformly uses the "IF-THEN" statement, and its general form is IF(LRC is LRCi and RRC is RRCj )THEN(Φf isΦfm and vb is vbn ). Among them, LRCi , RRCj , Φfm and vbn are fuzzy sets defined on the domain of discourse LRC , RRC , Φf and vb respectively.
上述“IF-THEN”模糊条件语句可归纳为控制规则R1～R16，每条规则的权重均为1，系统总的控制规则R可表示为模糊推理采用Mamdani方法，其中模糊关系蕴含运算采用Mamdani取小法，模糊关系合成运算采用Mamdani取大取小合成规则。选择Mamdani的原因在于其具有可直接应用于实际工程，参数设置简明，语言信息承载能力强等优势，方法成熟且应用广泛。The above "IF-THEN" fuzzy conditional statement can be summarized as control rules R1 ~ R16 , each rule has a weight of 1, and the total control rule R of the system can be expressed as The fuzzy reasoning adopts the Mamdani method, in which the implication operation of the fuzzy relation adopts the Mamdani small method, and the fuzzy relation composition operation adopts the Mamdani large and small composition rule. The reason for choosing Mamdani is that it has the advantages of being directly applicable to actual projects, concise parameter setting, and strong language information carrying capacity. The method is mature and widely used.
例如，当LRC＝x时，由LRC隶属度函数可得对于集合{Z,S,M,B}的隶属度μZ(LRC＝x)，μS(LRC＝x)，μM(LRC＝x)和μB(LRC＝x)；同理可得RRC＝y时对应于集合{Z,S,M,B}的隶属度μZ(RRC＝y)，μS(RRC＝y)、μM(RRC＝z)和μB(RRC＝z)。For example, when LRC =x, the membership degree μZ (LRC =x), μS (LRC =x), μ for the set {Z, S, M, B} can be obtained from the LRC membership functionM (LRC ＝x) and μB (LRC ＝x); similarly, when RRC =y, the degree of membership μZ (RRC ＝y) corresponding to the set {Z, S, M, B}, μS (RRC =y), μM (RRC =z) and μB (RRC =z).
表1模糊控制规则表Table 1 Fuzzy control rule table
Tab.1 Regulatory table of fuzzy controlTab.1 Regulatory table of fuzzy control
3.4去模糊化3.4 Defuzzification
FORD-FWMR驱动机构的输入为左、右前轮转角Φl和Φr以及左、右后轮转速nl和nr，但模糊推理机输出的前轮转角Φf和后轮转速vb均为模糊量，无法直接计算驱动输入，需要利用去模糊化步骤先将模糊推理机的输出转化为精确数值。常用的去模糊化方法包括面积重心法、加权平均法、取中位数法和平均最大隶属度法等。本文选择面积重心法实现去模糊化，具体过程如下：The inputs of the FORD-FWMR drive mechanism are the left and right front wheel rotation angles Φl and Φr and the left and right rear wheel speeds nl and nr , but the front wheel rotation angle Φf and the rear wheel speed vb output by the fuzzy reasoning machine are both is a fuzzy quantity, the driving input cannot be directly calculated, and the output of the fuzzy inference engine needs to be converted into an accurate value by using the defuzzification step. Commonly used defuzzification methods include area center of gravity method, weighted average method, median method and average maximum degree of membership method, etc. In this paper, the area center of gravity method is selected to achieve defuzzification. The specific process is as follows:
式(12)和式(13)中的∫表示输出模糊子集所有元素的隶属度值在连续论域上的代数积分，μΦ、μv分别表示前轮转角和行驶速度期望值的隶属度函数。∫ in Equation (12) and Equation (13) represents the algebraic integral of the membership degree values of all elements of the output fuzzy subset on the continuous universe, and μΦ and μv represent the membership function of the front wheel angle and the expected value of driving speed respectively .
4、基于运动学模型和模糊控制器输出项生成驱动机构输入信息，完成避障控制4. Based on the kinematics model and the output of the fuzzy controller, the input information of the driving mechanism is generated to complete the obstacle avoidance control
去模糊化完成后，利用得到的精确输出量结合公式(1)、(3)、(6)、(7)计算出下一时刻左、右前轮的转角和左、右后轮的转速，并转化为控制信号输入到驱动机构，从而实现对FORD-FWMR机器人的模糊避障决策与控制。After the defuzzification is completed, use the obtained accurate output combined with formulas (1), (3), (6), and (7) to calculate the rotation angle of the left and right front wheels and the speed of the left and right rear wheels at the next moment, And convert it into a control signal and input it to the drive mechanism, so as to realize the fuzzy obstacle avoidance decision and control of the FORD-FWMR robot.
综上所述，本发明主要目的为：针对前轮导向、后轮驱动的四轮移动机器人在障碍物密集环境中的智能避障需求，利用包含障碍物距离、尺寸、相对运动角度和相对速度等各种避障相关要素的碰撞风险模型来提高后续模糊避障决策与控制的性能，同时利用数量非常有限(2个)的模糊输入项降低模糊控制器的规则数量和计算成本，从而减小其响应时间和应用成本，实现在精简模糊规则数量前提下的较为完备、精细的避障决策与控制，对于改进前轮导向、后轮驱动的四轮移动机器人在仓储搬运、果园或温室作业等实际应用场景中的避障控制提供算法和技术参考。In summary, the main purpose of the present invention is to address the intelligent obstacle avoidance requirements of a front-wheel-guided, rear-wheel-driven four-wheel mobile robot in an obstacle-intensive environment, using The collision risk model of various obstacle avoidance-related elements is used to improve the performance of subsequent fuzzy obstacle avoidance decision-making and control. At the same time, a very limited number (2) of fuzzy input items is used to reduce the number of rules and calculation costs of the fuzzy controller, thereby reducing Its response time and application cost can realize relatively complete and fine obstacle avoidance decision-making and control under the premise of reducing the number of fuzzy rules. Obstacle avoidance control in practical application scenarios provides algorithm and technical reference.
进一步，为了在严格控制模糊规则数量的同时实现良好的避障控制性能，本算法针对FORD-FWMR在障碍物密集的未知环境下的自主避障需求，提出了一种基于机器人与障碍物发生碰撞的风险分析的模糊避障算法。该算法建立同时包含障碍物距离、尺寸、相对运动角度和相对速度等各种避障相关要素的碰撞风险模型，碰撞风险模型的计算主要涉及3个1次线性分段函数的加法和乘法，计算量很低。首先计算机器人左前方、右前方的碰撞风险，然后以计算结果为输入，生成一个双输入的模糊控制器。在两个输入项的模糊语言集合元素({零，小，中，大})的数量同时达到区分度较高的4时，整个模糊控制规则仍然只有16个，低于彭玉青的算法在输入区分度较低条件下的24个。Furthermore, in order to achieve good obstacle avoidance control performance while strictly controlling the number of fuzzy rules, this algorithm aims at the autonomous obstacle avoidance requirements of FORD-FWMR in an unknown environment with dense obstacles. A fuzzy obstacle avoidance algorithm for risk analysis. This algorithm establishes a collision risk model that includes various obstacle avoidance related elements such as obstacle distance, size, relative motion angle, and relative speed. The calculation of the collision risk model mainly involves the addition and multiplication of three linear piecewise functions. Volume is very low. Firstly, the collision risk of the left front and right front of the robot is calculated, and then a double-input fuzzy controller is generated with the calculation results as input. When the number of fuzzy language set elements ({zero, small, medium, large}) of the two input items reaches 4 with a high degree of discrimination at the same time, the entire fuzzy control rule is still only 16, which is lower than that of Peng Yuqing's algorithm in the input 24 in the less discriminative condition.
以上公开的仅为本发明的几个具体实施例，本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围，倘若本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。The above disclosures are only a few specific embodiments of the present invention, and those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention, provided that these modifications and modifications of the present invention belong to the rights of the present invention The present invention also intends to include these modifications and variations within the scope of the requirements and their technical equivalents.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910482568.4ACN110196596B (en) | 2019-06-04 | 2019-06-04 | Four-wheel mobile robot fuzzy obstacle avoidance method based on collision risk analysis |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910482568.4ACN110196596B (en) | 2019-06-04 | 2019-06-04 | Four-wheel mobile robot fuzzy obstacle avoidance method based on collision risk analysis |
| Publication Number | Publication Date |
|---|---|
| CN110196596Atrue CN110196596A (en) | 2019-09-03 |
| CN110196596B CN110196596B (en) | 2022-05-27 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910482568.4AActiveCN110196596B (en) | 2019-06-04 | 2019-06-04 | Four-wheel mobile robot fuzzy obstacle avoidance method based on collision risk analysis |
| Country | Link |
|---|---|
| CN (1) | CN110196596B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110906934A (en)* | 2019-11-29 | 2020-03-24 | 华中科技大学 | Unmanned ship obstacle avoidance method and system based on collision risk coefficient |
| CN112462786A (en)* | 2020-12-03 | 2021-03-09 | 大连海事大学 | Unmanned ship collision avoidance method based on fuzzy control strategy double-window algorithm |
| CN113204235A (en)* | 2021-04-13 | 2021-08-03 | 常州工学院 | Automatic obstacle avoidance algorithm of fire extinguishing robot |
| CN118387135A (en)* | 2024-04-19 | 2024-07-26 | 广东汽车检测中心有限公司 | A method for detecting and controlling automatic driving of an intelligent networked vehicle and a control system thereof |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101354587A (en)* | 2008-09-04 | 2009-01-28 | 湖南大学 | A multi-behavior fusion automatic navigation method for mobile robots in unknown environments |
| KR20110060711A (en)* | 2009-11-30 | 2011-06-08 | 재단법인대구경북과학기술원 | Obstacle Avoidance Fuzzy Controller for Mobile Robots |
| KR20120048351A (en)* | 2010-11-05 | 2012-05-15 | 재단법인대구경북과학기술원 | Method for remote control of mobile robot using fuzzy obstacle avoidance and system thereof |
| CN105116881A (en)* | 2015-06-26 | 2015-12-02 | 武汉科技大学 | Wheel-type robot self-adaptive navigation method based on on-line learning mechanism |
| CN108762256A (en)* | 2018-05-04 | 2018-11-06 | 中山大学 | The method of relatively high speed barrier is evaded by a kind of robot |
| CN109343346A (en)* | 2018-10-09 | 2019-02-15 | 浙江工业大学 | A method for automatic obstacle avoidance of self-balancing vehicle based on fuzzy control |
| CN109477723A (en)* | 2016-06-10 | 2019-03-15 | 杜克大学 | Autonomous Vehicle Motion Planning and Reconfigurable Motion Planning Processor |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101354587A (en)* | 2008-09-04 | 2009-01-28 | 湖南大学 | A multi-behavior fusion automatic navigation method for mobile robots in unknown environments |
| KR20110060711A (en)* | 2009-11-30 | 2011-06-08 | 재단법인대구경북과학기술원 | Obstacle Avoidance Fuzzy Controller for Mobile Robots |
| KR20120048351A (en)* | 2010-11-05 | 2012-05-15 | 재단법인대구경북과학기술원 | Method for remote control of mobile robot using fuzzy obstacle avoidance and system thereof |
| CN105116881A (en)* | 2015-06-26 | 2015-12-02 | 武汉科技大学 | Wheel-type robot self-adaptive navigation method based on on-line learning mechanism |
| CN109477723A (en)* | 2016-06-10 | 2019-03-15 | 杜克大学 | Autonomous Vehicle Motion Planning and Reconfigurable Motion Planning Processor |
| CN108762256A (en)* | 2018-05-04 | 2018-11-06 | 中山大学 | The method of relatively high speed barrier is evaded by a kind of robot |
| CN109343346A (en)* | 2018-10-09 | 2019-02-15 | 浙江工业大学 | A method for automatic obstacle avoidance of self-balancing vehicle based on fuzzy control |
| Title |
|---|
| 谢敬等: "基于碰撞危险度的移动机器人避障模糊控制", 《制造技术自动化学术会议论文集》* |
| 谢敬等: "基于碰撞危险度的移动机器人避障模糊控制", 《组合机床与自动化加工技术》* |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110906934A (en)* | 2019-11-29 | 2020-03-24 | 华中科技大学 | Unmanned ship obstacle avoidance method and system based on collision risk coefficient |
| CN112462786A (en)* | 2020-12-03 | 2021-03-09 | 大连海事大学 | Unmanned ship collision avoidance method based on fuzzy control strategy double-window algorithm |
| CN112462786B (en)* | 2020-12-03 | 2024-01-19 | 大连海事大学 | Unmanned ship collision prevention method based on fuzzy control strategy double-window algorithm |
| CN113204235A (en)* | 2021-04-13 | 2021-08-03 | 常州工学院 | Automatic obstacle avoidance algorithm of fire extinguishing robot |
| CN118387135A (en)* | 2024-04-19 | 2024-07-26 | 广东汽车检测中心有限公司 | A method for detecting and controlling automatic driving of an intelligent networked vehicle and a control system thereof |
| Publication number | Publication date |
|---|---|
| CN110196596B (en) | 2022-05-27 |
| Publication | Publication Date | Title |
|---|---|---|
| CN110196596B (en) | Four-wheel mobile robot fuzzy obstacle avoidance method based on collision risk analysis | |
| CN114407931B (en) | A highly human-like safe driving decision-making method for autonomous driving commercial vehicles | |
| Yu et al. | RDT-RRT: Real-time double-tree rapidly-exploring random tree path planning for autonomous vehicles | |
| CN100491084C (en) | A local path planning method for mobile robots based on binary environment information | |
| CN101833333B (en) | Unknown environment route planning method of underwater cleaning robot | |
| CN106909151A (en) | For the unpiloted path planning of agricultural machinery and its control method | |
| CN105335797B (en) | A Trajectory Optimization Method for Autonomous Parking Based on Fully Simultaneous Dynamic Optimization Framework | |
| CN103439884B (en) | A method of intelligent vehicle lateral control based on fuzzy sliding mode | |
| Mester | Intelligent Mobile Robot Controller Design | |
| Mac et al. | MIMO fuzzy control for autonomous mobile robot | |
| CN111538328B (en) | A priority hierarchical predictive control method for obstacle avoidance trajectory planning and tracking control of autonomous vehicles | |
| Li et al. | Path planning and obstacle avoidance control for autonomous multi-axis distributed vehicle based on dynamic constraints | |
| CN111522337A (en) | A multi-drive wheel AGV navigation method based on fuzzy control | |
| CN102609765A (en) | Intelligent vehicle lane change path planning method based on polynomial and radial basis function (RBF) neural network | |
| CN107168309A (en) | A kind of underwater multi-robot paths planning method of Behavior-based control | |
| CN109960150A (en) | Tracking control method of inspection robot | |
| CN110471289A (en) | A kind of the Adaptive Path tracking and system of view-based access control model navigation mobile device | |
| Mester | Intelligent mobile robot control in unknown environments | |
| Yang et al. | Sobol sequence RRT* and numerical optimal joint algorithm-based automatic parking trajectory planning of four-wheel steering vehicles | |
| Malhotra et al. | Development of a fuzzy logic based mobile robot for dynamic obstacle avoidance and goal acquisition in an unstructured environment | |
| Si et al. | Path planning of factory handling robot integrating fuzzy logic-PID control technology | |
| CN114859918A (en) | Intelligent vehicle trajectory tracking control system and method based on interval type 2 fuzzy logic | |
| Song et al. | A TC-RRT-based path planning algorithm for the nonholonomic mobile robots | |
| CN118518105A (en) | Based on improve A*Multi-target plant protection robot obstacle avoidance path optimization method of-IWOA | |
| CN118466209A (en) | Parameter adaptive MPC path tracking control method for rear-wheel steering agricultural machinery |
| 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 |