WO2020133887A1 - Industrial robot collision prevention method - Google Patents
Industrial robot collision prevention method Download PDFInfo
- Publication number
- WO2020133887A1 WO2020133887A1 PCT/CN2019/087243 CN2019087243W WO2020133887A1 WO 2020133887 A1 WO2020133887 A1 WO 2020133887A1 CN 2019087243 W CN2019087243 W CN 2019087243W WO 2020133887 A1 WO2020133887 A1 WO 2020133887A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- robot
- workpiece
- joint
- collision
- bounding box
- Prior art date
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J19/00—Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
- B25J19/06—Safety devices
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
Definitions
- the invention relates to industrial robots, in particular to an anti-collision method for industrial robots.
- the existing collision detection methods mainly include offline simulation based on path planning and real-time detection based on torque feedback.
- the offline detection method often requires a lot of computing resources; the robot running program needs to be written in advance; and because it is only for the pre-configured workspace, the collision model cannot be modified according to the changes in real-time working conditions, and it cannot be jogged and followed on the robot.
- Collision detection in working mode The real-time detection method based on torque feedback is mainly based on analyzing the abnormal torque fluctuations at the time of collision to achieve the purpose of collision detection and timely stop loss and stop. This kind of method cannot substantially avoid the occurrence of collision, and is often caused by the high detection sensitivity. False alarm.
- Chinese patent CN103192413A discloses a sensorless robot collision detection device and method.
- the calculation module calculates the difference between the robot motion state and the actual planned state of the robot at the next moment predicted by the observation module; the comparison of the judgment module Whether the difference exceeds the set threshold; when the difference is greater than the set threshold, the execution module drives the robot to stop.
- Neural network training in this method requires a large number of samples and time cost, and is not suitable for industrial robots in flexible manufacturing systems that need to frequently change tools, and this method only stops the robot when the robot collides, and cannot prevent collisions.
- Chinese patent CN101719173A discloses an object-oriented parallel collision detection method and system.
- the system consists of a main control node and a detection node.
- the main control point is responsible for allocating detection tasks, and the detection node is responsible for detecting whether the assigned pair of objects collide.
- the object model proposed in this method is an arbitrary polyhedron. In an actual complex production environment, this object model will increase the time complexity of the collision detection algorithm of a single node, and each cycle needs to detect all moving objects and all other Whether the object collides, the complexity of the detection algorithm increases.
- the invention provides an anti-collision method for industrial robots, which does not need any additional external sensors, and does not modify the mechanical structure of the robot. It is suitable for robot teaching, reproduction, following, and automatic running states, without the need for simulation testing with other software.
- An anti-collision method for industrial robots includes the following steps connected in sequence:
- Step 1 Establish the geometric model of the robot, workpiece and external tooling in the initial state, and simplify the AABB bounding box model incorporating external tooling;
- Step 2 When the robot is moving, in the first control cycle, predict the position and posture of each joint, rod and workpiece of the robot after m cycles according to the observed robot motion state, and based on the predicted position of each joint, rod and workpiece of the robot And posture refresh the geometric model of robot and workpiece;
- Step 3 Check whether the geometric models of the joints and rods of the workpiece and the robot body intersect; if they intersect, the workpiece and the robot body will be collided and stop the robot immediately; if they do not intersect, continue the collision detection between the robot, the workpiece and the external tooling;
- Step 4 According to the position and posture of the robot joints, rods, and workpieces after the prediction of m cycles, generate a robot-workpiece AABB bounding box model, and check whether the robot-workpiece AABB bounding box model and the combined external tooling AABB bounding box model Intersect; if it does not intersect, it is considered that there is no potential collision risk; if it intersects, screen the intersection area and extract the list of external tooling to be tested;
- Step 5 Check whether the geometric model of the robot and the workpiece intersects the geometric model of the external tooling in the external tooling list to be tested; if it intersects, it is considered that a collision occurs, and the robot is stopped immediately; if it does not intersect, it is considered that there is no potential collision risk; continue to the next control cycle;
- Step 6 Repeat steps 2 to 5 in the next control cycle until the robot completes all actions.
- the geometric model of the robot and the workpiece is a collection of three-dimensional figures composed of cubes and spheres generated according to the mechanical attributes of the robot and the workpiece; the geometric model defined by the mechanical attributes of the external tooling is to teach the external by operating the robot The vertex of the tooling, the cube model generated by inputting the length, width and height.
- the method for generating the simplified AABB bounding box model of merging external tooling in step one is to merge the geometric models defined by the mechanical properties of the adjacent external tooling in a teaching area to generate an AABB enclosing containing multiple external toolings Box model.
- step 2 the calculation method of the m period described in step 2 and step 2 is:
- ⁇ is the compensation period parameter.
- the introduction of compensation cycle parameters can improve the flexibility of the collision detection algorithm.
- step 2 the method for refreshing the geometric model of the robot and the workpiece according to the predicted positions and postures of each joint, rod and workpiece of the robot in step 2 is:
- the method for generating the robot-workpiece AABB bounding box model described in step 4 is: obtaining the updated robot model by updating the robot motion state, and obtaining the vertices of the robot and the workpiece in the x, y, and z directions based on the base coordinate system To generate the AABB bounding box model.
- the anti-collision method for industrial robots proposed in the present invention does not require any additional external sensors, and does not modify the mechanical structure of the robot, and is applicable to any operating conditions of the robot.
- the anti-collision method for industrial robots proposed by the present invention is suitable for robot teaching, reproducing, following, and automatic running states, without the need to use other software for simulation testing.
- the industrial robot anti-collision method proposed by the present invention first refreshes the geometric model of the robot and the workpiece according to the predicted robot motion state, and detects the intersection of the geometric model of the robot workpiece to detect whether the workpiece has the risk of collision with the robot body; detection After the robot workpiece has no risk of collision, the AABB bounding box model is established according to the geometric model of the refreshed robot and the workpiece, and the intersection detection with the simplified workpiece AABB bounding box model. Finally, only the robot, the workpiece and the external tooling in the intersection area are detected, not in the intersection area. The tooling inside does not need to be intersected to reduce the amount of intersection detection. The procedure is simple and the execution efficiency is high.
- the industrial robot anti-collision method proposed by the present invention can detect the collision risk in time before a potential collision occurs, and stop the robot to avoid the collision.
- FIG. 1 is a system flowchart of Embodiment 1 of the industrial robot anti-collision method of the present invention.
- FIG. 2 is a schematic diagram of a robot-workpiece AABB bounding box model of Embodiment 1 of the industrial robot anti-collision method of the present invention.
- FIG. 3 is a model diagram of a combined and simplified external tooling AABB bounding box of Embodiment 1 of the industrial robot anti-collision method of the present invention.
- Embodiment 1 is a method in which an industrial robot composed of a base, an arm, an arm, and a wrist clamps a workpiece, and detects whether the industrial robot collides with the clamped workpiece and external tooling.
- the robot anti-collision method of Embodiment 1 includes the following steps connected in sequence:
- Step 1.1 Establish the robot geometric model, workpiece geometric model and external tooling geometric model in the initial state
- the geometric model of the robot and the workpiece is a geometric model defined by the mechanical properties of the robot and the workpiece, that is, a three-dimensional graphic set composed of cubes and spheres generated according to the mechanical properties of the robot and the workpiece.
- the external tooling geometric model is a cube model generated by inputting the length, width and height by teaching the vertices of the external tooling by operating the robot.
- Step 1.2 based on the establishment of the external tooling geometric model, simplify the combination of external tooling to generate the AABB bounding box model
- the method is to merge the geometric models defined by the mechanical properties of the adjacent external tooling by means of the teaching area to generate an AABB bounding box model containing multiple external toolings.
- external tool 1 external tool 2
- external tool 3 external tool 3.
- the generated AABB bounding box model of the external tool is shown in FIG. 3.
- Step 2 The robot moves, in the first control cycle, predicts the position and posture of the robot joints, rods, and workpieces after m cycles based on the observed robot motion state, and the predicted position and position of the robot joints, rods, and workpieces after m cycles.
- the robot joints are the joints of the base and the boom, the joints of the boom and the forearm, and the joints of the forearm and the wrist, which are set to joint 1, joint 2, and joint 3.
- the robot rods are the big arm, the forearm and the wrist, and they are set as the rod 1, the rod 2 and the rod 3.
- the controller calculates the time required for each joint to stop at the maximum speed that is, the maximum stop time of each joint, and the joint stop time with the longest stop time is selected as the maximum time for robot motion to stop t stop ; the controller calculation cycle ⁇ t is obtained, and the predicted cycle number is calculated;
- the method of refreshing the geometric model of the robot and the workpiece according to the predicted positions and postures of each joint, rod and workpiece of the robot is:
- Step 3 Check whether the geometric models of the joints and rods of the workpiece and the robot body intersect; if they intersect, the workpiece and the robot body will collide and stop the robot immediately; if they do not intersect, continue the collision detection of the robot, the workpiece and the external tooling;
- Step 4 According to the predicted positions and postures of the robot joints, rods, and workpieces after 5 cycles, generate a robot-workpiece AABB bounding box model, and check whether the robot-workpiece AABB bounding box model and the combined external tooling AABB bounding box model Intersect; if it does not intersect, it is considered that there is no potential collision risk; if it intersects, screen the intersecting area and bring up the list of external tooling to be tested;
- the method of generating a robot-workpiece AABB bounding box model is to obtain the updated robot model by updating the robot motion state, obtain the vertices of the robot and the workpiece in the x, y, and z directions based on the base coordinate system, and generate the AABB bounding box model. See Figure 2.
- Step 5 Check whether the geometric model of the robot and the workpiece intersect with the external tool geometric model in the external tool list to be tested; if it intersects, it is considered that a collision has occurred, and the robot is stopped immediately; if it does not intersect, it is considered that there is no potential collision risk; continue to the next control cycle;
- Step 6 Repeat Step 2 to Step 5 in the next control cycle until the robot completes all actions.
- ⁇ the smaller the collision detection sensitivity, that is, the shorter the distance between the robot's stopping position and the potential collision external tooling after detecting the potential collision risk.
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Manipulator (AREA)
- Numerical Control (AREA)
Abstract
Provided is an industrial robot collision prevention method. Firstly, geometric models of a robot and a workpiece are refreshed according to predicted robot motion states, and an intersection test is performed on the geometric models of the robot and the workpiece, so as to detect whether there is a risk of the workpiece colliding with the robot body. Upon detecting there is no risk of the workpiece and the robot colliding, an AABB bounding box model is established according to the refreshed geometric models of the robot and the workpiece, and an intersection test is performed on the AABB bounding box model and a simplified workpiece AABB bounding box model. Finally, an intersection test is performed on the robot, the workpiece, and external tools in an intersection region without subjecting tools not present in the intersection region to the intersection test, such that operations of the intersection test are reduced. The method detects a collision risk in a timely manner to stop robot motion and prevent a potential collision before the collision occurs. The method is applicable to any operational conditions of the robot.
Description
本发明涉及工业机器人，具体涉及一种工业机器人防碰撞方法。The invention relates to industrial robots, in particular to an anti-collision method for industrial robots.
工业机器人在运动过程中，因各种原因将导致机器人本体与夹具、机器人与外部工装或人员发生碰撞，致使机器人本体、工件和外部工装受到损坏或损伤。因此如何检测及避免碰撞在机器人应用中至关重要。During the movement of the industrial robot, due to various reasons, the robot body and the fixture, the robot and the external tooling or personnel will collide, causing the robot body, workpiece and external tooling to be damaged or damaged. Therefore, how to detect and avoid collisions is very important in robot applications.
现有碰撞检测方法主要有基于路径规划的离线仿真以及基于力矩反馈的实时检测两种方法。离线检测的方法往往需要大量计算资源；需要预先编写好机器人运行程序；且由于仅针对预先配置好的工作空间，无法根据实时工况的改变进行碰撞模型的修正，并且无法在机器人点动、跟随工作模式下进行碰撞检测。基于力矩反馈的实时检测方法主要根据分析碰撞发生时的力矩异常波动以达到碰撞检测并及时止损和停止的目的，该类方法实质上无法避免碰撞的发生，且经常由于检测灵敏度过高而导致误报警。The existing collision detection methods mainly include offline simulation based on path planning and real-time detection based on torque feedback. The offline detection method often requires a lot of computing resources; the robot running program needs to be written in advance; and because it is only for the pre-configured workspace, the collision model cannot be modified according to the changes in real-time working conditions, and it cannot be jogged and followed on the robot. Collision detection in working mode. The real-time detection method based on torque feedback is mainly based on analyzing the abnormal torque fluctuations at the time of collision to achieve the purpose of collision detection and timely stop loss and stop. This kind of method cannot substantially avoid the occurrence of collision, and is often caused by the high detection sensitivity. False alarm.
中国专利CN103192413A公开一种无传感器的机器人碰撞检测装置及方法，由计算模组根据观测模组预测的下一时刻机器人运动状态计算出其与机器人实际规划状态之间的差值；判断模组对比该差值是否超过所设阈值；当差值大于所设阈值，执行模组驱使机器人停止。该方法中神经网络的训练需要大量的样本和时间成本，不适用于柔性制造***中需要频繁更换工具的工业机器人，并且该方法仅仅在机器人发生碰撞时停止机器人，无法预防碰撞的发生。Chinese patent CN103192413A discloses a sensorless robot collision detection device and method. The calculation module calculates the difference between the robot motion state and the actual planned state of the robot at the next moment predicted by the observation module; the comparison of the judgment module Whether the difference exceeds the set threshold; when the difference is greater than the set threshold, the execution module drives the robot to stop. Neural network training in this method requires a large number of samples and time cost, and is not suitable for industrial robots in flexible manufacturing systems that need to frequently change tools, and this method only stops the robot when the robot collides, and cannot prevent collisions.
中国专利CN101719173A公开的一种面向对象的碰撞并行检测方法及***。该***由主控制节点和检测节点组成，主控制点负责分配检测任务，检测节点负责检测分配的一对物体是否碰撞。该方法中提出的物体模型为任意多面体，这种物体模型在实际复杂的生产环境中，会导致单个节点的碰撞检测算法的时间复杂度增加，并且每个周期都需要检测所有运动物体与所有其他物体是否碰撞，检测算法的复杂度增加。Chinese patent CN101719173A discloses an object-oriented parallel collision detection method and system. The system consists of a main control node and a detection node. The main control point is responsible for allocating detection tasks, and the detection node is responsible for detecting whether the assigned pair of objects collide. The object model proposed in this method is an arbitrary polyhedron. In an actual complex production environment, this object model will increase the time complexity of the collision detection algorithm of a single node, and each cycle needs to detect all moving objects and all other Whether the object collides, the complexity of the detection algorithm increases.
发明内容Summary of the invention
本发明提供一种工业机器人防碰撞方法，无需附加任何外部传感器，且对机器人机械结构无任何修改，适用于机器人示教、再现、跟随、自动运行状态，无需借助其他软件进行仿真测试。The invention provides an anti-collision method for industrial robots, which does not need any additional external sensors, and does not modify the mechanical structure of the robot. It is suitable for robot teaching, reproduction, following, and automatic running states, without the need for simulation testing with other software.
为此，本发明采用的技术方案是：For this reason, the technical solutions adopted by the present invention are:
一种工业机器人防碰撞方法，包括顺次相接的以下步骤：An anti-collision method for industrial robots includes the following steps connected in sequence:
步骤一，分别建立初始状态下机器人、工件及外部工装的几何模型，并简化合并外部工装的AABB包围盒模型；Step 1: Establish the geometric model of the robot, workpiece and external tooling in the initial state, and simplify the AABB bounding box model incorporating external tooling;
步骤二，机器人动作时，在第一控制周期内，根据观测的机器人运动状态预测m周期后 机器人各关节、杆件、工件的位置和姿态，根据预测的机器人各关节、杆件和工件的位置和姿态刷新机器人、工件的几何模型；Step 2: When the robot is moving, in the first control cycle, predict the position and posture of each joint, rod and workpiece of the robot after m cycles according to the observed robot motion state, and based on the predicted position of each joint, rod and workpiece of the robot And posture refresh the geometric model of robot and workpiece;
步骤三，检测工件与机器人本体各关节、杆件几何模型是否相交；若相交，认为工件与机器人本体将发生碰撞，立即停止机器人；若不相交，继续机器人、工件与外部工装碰撞检测；Step 3: Check whether the geometric models of the joints and rods of the workpiece and the robot body intersect; if they intersect, the workpiece and the robot body will be collided and stop the robot immediately; if they do not intersect, continue the collision detection between the robot, the workpiece and the external tooling;
步骤四，根据预测m周期后机器人各关节、杆件、工件的位置和姿态，生成机器人-工件的AABB包围盒模型，检测机器人-工件的AABB包围盒模型与合并外部工装的AABB包围盒模型是否相交；若不相交，认为无潜在碰撞风险发生；若相交，筛选相交区域，提取待检测外部工装列表；Step 4: According to the position and posture of the robot joints, rods, and workpieces after the prediction of m cycles, generate a robot-workpiece AABB bounding box model, and check whether the robot-workpiece AABB bounding box model and the combined external tooling AABB bounding box model Intersect; if it does not intersect, it is considered that there is no potential collision risk; if it intersects, screen the intersection area and extract the list of external tooling to be tested;
步骤五，检测机器人、工件几何模型与待检测外部工装列表中的外部工装几何模型是否相交；若相交，认为发生碰撞，立即停止机器人；若不相交，认为无潜在碰撞风险发生；继续下一个控制周期；Step 5: Check whether the geometric model of the robot and the workpiece intersects the geometric model of the external tooling in the external tooling list to be tested; if it intersects, it is considered that a collision occurs, and the robot is stopped immediately; if it does not intersect, it is considered that there is no potential collision risk; continue to the next control cycle;
步骤六，下一个控制周期重复步骤2至步骤5，直到机器人完成全部动作。Step 6: Repeat steps 2 to 5 in the next control cycle until the robot completes all actions.
进一步地，步骤一所述机器人、工件的几何模型为根据机器人、工件机械属性生成的由立方体和球体组成的立体图形的集合；由外部工装的机械属性定义的几何模型为通过操作机器人示教外部工装顶点，通过输入长宽高生成的立方体模型。Further, in step 1, the geometric model of the robot and the workpiece is a collection of three-dimensional figures composed of cubes and spheres generated according to the mechanical attributes of the robot and the workpiece; the geometric model defined by the mechanical attributes of the external tooling is to teach the external by operating the robot The vertex of the tooling, the cube model generated by inputting the length, width and height.
进一步地，步骤一所述简化合并外部工装的AABB包围盒模型的生成方法为：通过示教区域的方式将临近的外部工装的机械属性定义的几何模型合并，生成包含多个外部工装的AABB包围盒模型。Further, the method for generating the simplified AABB bounding box model of merging external tooling in step one is to merge the geometric models defined by the mechanical properties of the adjacent external tooling in a teaching area to generate an AABB enclosing containing multiple external toolings Box model.
进一步地，步骤二步骤二所述m周期的计算方法为：Further, the calculation method of the m period described in step 2 and step 2 is:
获取在本周期内的机器人各关节最大速度
以及最大加速度
计算各关节在最大速度下停止所需时间
即各关节最大停止时间，选取停止时间最长的关节停止时间为机器人动作停止最长时间t
stop；获取控制器运算周期Δt，计算预测周期数；
Get the maximum speed of each joint of the robot in this cycle And maximum acceleration Calculate the time required for each joint to stop at the maximum speed That is, the maximum stop time of each joint, and the joint stop time with the longest stop time is selected as the maximum time for robot motion to stop t stop ; the controller calculation cycle Δt is obtained, and the predicted cycle number is calculated;
m＝t
stop/Δt。
m=t stop /Δt.
进一步地，在确认预测周期m的方法中引入补偿周期参数γ，m＝t
stop/Δt+γ；其中γ为补偿周期参数。根据机器人运行工况不同，适当增大或减少γ。γ值越小，则碰撞预测距离越短，碰撞检测灵敏度越小，即检测到潜在碰撞风险后机器人停止位置与潜在碰撞的外部工装之间距离越近。引入补偿周期参数，可提高碰撞检测算法的灵活度。
Further, the compensation period parameter γ is introduced into the method for confirming the prediction period m, m=t stop /Δt+γ; where γ is the compensation period parameter. According to the different operating conditions of the robot, increase or decrease γ appropriately. The smaller the γ value, the shorter the collision prediction distance and the smaller the collision detection sensitivity, that is, the closer the distance between the robot stop position and the potential collision external tooling after detecting the potential collision risk. The introduction of compensation cycle parameters can improve the flexibility of the collision detection algorithm.
进一步地，步骤二所述根据预测的机器人各关节、杆件和工件的位置和姿态刷新机器人、工件的几何模型的方法为：Further, the method for refreshing the geometric model of the robot and the workpiece according to the predicted positions and postures of each joint, rod and workpiece of the robot in step 2 is:
⑴、从控制器中获取当前时刻机器人各关节位置θ
1,θ
2,θ
3…θ
i，关节速度
(1) Obtain the robot joint positions θ 1 , θ 2 , θ 3 …θ i , joint velocity from the controller at the current time
⑵、使用雅可比方程：
计算各关节和工件在坐标系下合成速度V
i，其中v
i(t)为关节线速度，ω
i(t)为t时刻关节姿态速度；
⑵、Use Jacobian equation: Calculate the combined velocity V i of each joint and workpiece in the coordinate system, where v i (t) is the joint linear velocity, and ω i (t) is the joint posture velocity at t;
⑶、设定关节在当前时刻t做匀速直线运动，通过状态方程计算m周期后各关节、杆件和工件几何模型位置和姿态；(3) Set the joints to perform linear movement at a constant speed at the current time t, and calculate the position and posture of the geometric models of each joint, rod and workpiece after the m cycle through the state equation;
P
i(t+m·Δt)＝v
i(t)·m·Δt+P
i(t)
P i (t+m·Δt)=v i (t)·m·Δt+P i (t)
O
i(t+m·Δt)＝ω
i(t)·m·Δt+O
i(t)
O i (t+m·Δt)=ω i (t)·m·Δt+O i (t)
其中Δt为控制器运算周期，P
i(t)为控制器运算周期，Δt为控制器运算周期，P
i(t)为当前时刻t下各关节、杆件和工件几何模型位置，O
i(t)为当前时刻t下各关节、杆件和工件几何模型姿态，计算公式为：
Where Δt is the controller calculation cycle, P i (t) is the controller calculation cycle, Δt is the controller calculation cycle, P i (t) is the position of the geometric model of each joint, rod and workpiece at the current time t, O i ( t) is the geometric model pose of each joint, rod and workpiece at the current time t, the calculation formula is:
⑷、根据机器人各关节，杆件和工件位置和姿态刷新机器人、工件几何模型。⑷ Refresh the geometric model of the robot and the workpiece according to the positions and postures of the robot joints, rods and workpieces.
进一步地，步骤四所述生成机器人-工件的AABB包围盒模型的方法为：通过更新机器人运动状态，获取更新的机器人模型，基于基坐标系获取机器人和工件各x、y、z方向上的顶点，生成AABB包围盒模型。Further, the method for generating the robot-workpiece AABB bounding box model described in step 4 is: obtaining the updated robot model by updating the robot motion state, and obtaining the vertices of the robot and the workpiece in the x, y, and z directions based on the base coordinate system To generate the AABB bounding box model.
本发明的有益效果：The beneficial effects of the invention:
1、本发明所提出的工业机器人防碰撞方法，无需附加任何外部传感器，且对机器人机械结构无任何修改，适用于机器人的任何运行工况。1. The anti-collision method for industrial robots proposed in the present invention does not require any additional external sensors, and does not modify the mechanical structure of the robot, and is applicable to any operating conditions of the robot.
2、本发明所提出的工业机器人防碰撞方法，适用于机器人示教、再现、跟随、自动运行状态，无需借助其他软件进行仿真测试。2. The anti-collision method for industrial robots proposed by the present invention is suitable for robot teaching, reproducing, following, and automatic running states, without the need to use other software for simulation testing.
3、本发明所提出的工业机器人防碰撞方法，首先根据预测的机器人运动状态刷新机器人和工件的几何模型，进行机器人工件几何模型相交检测，以检测工件是否存在与机器人本体发生碰撞的风险；检测机器人工件无碰撞风险后，再根据刷新机器人和工件的几何模型建立AABB包围盒模型，与简化工件AABB包围盒模型相交检测，最后仅将机器人、工件与相交区域内外部工装相交检测，不在相交区域内的工装无需进行相交检测，以减少相交检测算量。程序简单，执行效率高。3. The industrial robot anti-collision method proposed by the present invention first refreshes the geometric model of the robot and the workpiece according to the predicted robot motion state, and detects the intersection of the geometric model of the robot workpiece to detect whether the workpiece has the risk of collision with the robot body; detection After the robot workpiece has no risk of collision, the AABB bounding box model is established according to the geometric model of the refreshed robot and the workpiece, and the intersection detection with the simplified workpiece AABB bounding box model. Finally, only the robot, the workpiece and the external tooling in the intersection area are detected, not in the intersection area. The tooling inside does not need to be intersected to reduce the amount of intersection detection. The procedure is simple and the execution efficiency is high.
4、本发明所提出的工业机器人防碰撞方法，可在潜在的碰撞发生前及时检测出碰撞风险，并停止机器人，避免碰撞发生。4. The industrial robot anti-collision method proposed by the present invention can detect the collision risk in time before a potential collision occurs, and stop the robot to avoid the collision.
图1是本发明工业机器人防碰撞方法的实施方式1的***流程图。FIG. 1 is a system flowchart of Embodiment 1 of the industrial robot anti-collision method of the present invention.
图2是本发明工业机器人防碰撞方法的实施方式1的机器人-工件AABB包围盒模型示意图。2 is a schematic diagram of a robot-workpiece AABB bounding box model of Embodiment 1 of the industrial robot anti-collision method of the present invention.
图3是本发明工业机器人防碰撞方法的实施方式1的合并简化的外部工装AABB包围盒模型图。FIG. 3 is a model diagram of a combined and simplified external tooling AABB bounding box of Embodiment 1 of the industrial robot anti-collision method of the present invention.
为使本发明的目的、技术方案和优点更加清楚，下面将结合附图及实施方式1及2对本发明的技术方案进行清楚、完整地描述。To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described clearly and completely in conjunction with the drawings and Embodiments 1 and 2.
实施方式1Embodiment 1
实施方式1为一个由基座、大臂、小臂、手腕构成的工业机器人夹持工件运动，检测该工业机器人与夹持工件、外部工装是否相碰撞的方法。Embodiment 1 is a method in which an industrial robot composed of a base, an arm, an arm, and a wrist clamps a workpiece, and detects whether the industrial robot collides with the clamped workpiece and external tooling.
假定机器人夹持工件运动的当前运算时刻为t＝50ms，控制运算周期长为10ms；外部工装共有三个，即外部工装1、外部工装2及外部工装3。Assume that the current calculation time of the robot gripping the workpiece motion is t=50ms, and the control calculation period is 10ms; there are three external toolings, namely external tooling 1, external tooling 2, and external tooling 3.
参阅图1，实施方式1的机器人防碰撞方法包括顺次相接的以下步骤：Referring to FIG. 1, the robot anti-collision method of Embodiment 1 includes the following steps connected in sequence:
步骤1.1，建立初始状态下的机器人几何模型、工件几何模型及外部工装几何模型；Step 1.1: Establish the robot geometric model, workpiece geometric model and external tooling geometric model in the initial state;
机器人、工件的几何模型为由机器人、工件机械属性定义的几何模型，即根据机器人、工件机械属性生成的由立方体和球体组成的立体图形集合。外部工装几何模型为通过操作机器人示教外部工装顶点，通过输入长宽高生成的立方体模型。The geometric model of the robot and the workpiece is a geometric model defined by the mechanical properties of the robot and the workpiece, that is, a three-dimensional graphic set composed of cubes and spheres generated according to the mechanical properties of the robot and the workpiece. The external tooling geometric model is a cube model generated by inputting the length, width and height by teaching the vertices of the external tooling by operating the robot.
步骤1.2，在建立了外部工装几何模型的基础上，简化合并外部工装生成AABB包围盒模型；Step 1.2, based on the establishment of the external tooling geometric model, simplify the combination of external tooling to generate the AABB bounding box model;
方法为：通过示教区域的方式将临近的外部工装的机械属性定义的几何模型合并，生成包含多个外部工装的AABB包围盒模型。在本实施例中，假定有三个外部工装，即外部工装1、外部工装2及外部工装3，生成的外部工装的AABB包围盒模型如附图3所示。The method is to merge the geometric models defined by the mechanical properties of the adjacent external tooling by means of the teaching area to generate an AABB bounding box model containing multiple external toolings. In this embodiment, it is assumed that there are three external tools, namely, external tool 1, external tool 2, and external tool 3. The generated AABB bounding box model of the external tool is shown in FIG. 3.
步骤2，机器人动作，在第一控制周期内，根据观测的机器人运动状态预测m周期后机器人各关节、杆件、工件的位置和姿态，根据预测的机器人各关节、杆件和工件的位置和姿态刷新机器人、工件的几何模型；Step 2: The robot moves, in the first control cycle, predicts the position and posture of the robot joints, rods, and workpieces after m cycles based on the observed robot motion state, and the predicted position and position of the robot joints, rods, and workpieces after m cycles. Posture refreshing geometric model of robot and workpiece;
参见图2，在本实施例中，机器人关节为基座与大臂连接处、大臂与小臂连接处、及小臂与手腕连接处，设定其为关节1、关节2及关节3。机器人杆件为大臂、小臂及手腕，设定其为杆件1、杆件2及杆件3。Referring to FIG. 2, in this embodiment, the robot joints are the joints of the base and the boom, the joints of the boom and the forearm, and the joints of the forearm and the wrist, which are set to joint 1, joint 2, and joint 3. The robot rods are the big arm, the forearm and the wrist, and they are set as the rod 1, the rod 2 and the rod 3.
m周期的计算方法为：The calculation method of m period is:
根据机器人性能，即控制器中设置的机器人各关节运动参数中的最大速度
以及最大加速度
计算各关节在最大速度下停止所需时间
即各关节最大停止时间，选取停止时间最长的关节停止时间为机器人动作停止最长时间t
stop；获取控制器运算周期Δt，计算预测周期数；
According to the robot performance, that is, the maximum speed of the robot's joint motion parameters set in the controller And maximum acceleration Calculate the time required for each joint to stop at the maximum speed That is, the maximum stop time of each joint, and the joint stop time with the longest stop time is selected as the maximum time for robot motion to stop t stop ; the controller calculation cycle Δt is obtained, and the predicted cycle number is calculated;
m＝t
stop/Δt
m=t stop /Δt
假定关节3的最大速度为100°/s，最大加速度为2000°/s，则关节3的停止时间为
Assuming that the maximum speed of joint 3 is 100°/s and the maximum acceleration is 2000°/s, the stop time of joint 3 is
假定关节3的停止时间最长，为50ms，控制器运算周期为10ms，m＝50/10＝5。Assume that the stop time of the joint 3 is the longest, which is 50 ms, the controller calculation cycle is 10 ms, and m=50/10=5.
根据预测的机器人各关节、杆件和工件的位置和姿态刷新机器人、工件的几何模型的方法为：The method of refreshing the geometric model of the robot and the workpiece according to the predicted positions and postures of each joint, rod and workpiece of the robot is:
⑴、从控制器中获取当前时刻机器人各关节位置θ
1,θ
2,θ
3…θ
i，关节速度
(1) Obtain the robot joint positions θ 1 , θ 2 , θ 3 …θ i , joint velocity from the controller at the current time
⑵、使用雅可比方程：
计算各关节和工件在坐标系下合成速度V
i，其中v
i(50)为关节线速度，ω
i(50)为t时刻关节姿态速度；
⑵、Use Jacobian equation: Calculate the combined velocity V i of each joint and workpiece in the coordinate system, where v i (50) is the joint linear velocity and ω i (50) is the joint posture velocity at time t;
⑶、设定关节在当前时刻做匀速直线运动，通过状态方程计算5个周期后各关节、杆件和工件几何模型位置和姿态；(3) Set the joints to perform linear motion at a constant speed at the current moment, and calculate the position and posture of the geometric models of each joint, rod and workpiece after 5 cycles through the state equation;
P
i(50+m·10)＝v
i(t)·m·10+P
i(50)
P i (50+m·10)=v i (t)·m·10+P i (50)
O
i(50+m·10)＝ω
i(t)·m·10+O
i(50)
O i (50+m·10)=ω i (t)·m·10+O i (50)
其中P
i(50)为当前时刻第50ms下各关节、杆件和工件几何模型位置，O
i(50)为当前时刻第50ms下各关节、杆件和工件几何模型姿态，计算公式为：
Where P i (50) is the position of geometric models of joints, members and workpieces at the current time of 50ms, and O i (50) is the pose of geometric models of joints, members and workpieces at the current time of 50ms, and the calculation formula is:
⑷、根据机器人各关节，杆件和工件位置和姿态刷新机器人、工件几何模型。⑷ Refresh the geometric model of the robot and the workpiece according to the positions and postures of the robot joints, rods and workpieces.
步骤3，检测工件与机器人本体各关节、杆件几何模型是否相交；若相交，认为工件与机器人本体将发生碰撞，立即停止机器人；若不相交，继续机器人、工件与外部工装碰撞检测；Step 3: Check whether the geometric models of the joints and rods of the workpiece and the robot body intersect; if they intersect, the workpiece and the robot body will collide and stop the robot immediately; if they do not intersect, continue the collision detection of the robot, the workpiece and the external tooling;
步骤4，根据预测5周期后机器人各关节、杆件、工件的位置和姿态，生成机器人-工件的AABB包围盒模型，检测机器人-工件的AABB包围盒模型与合并外部工装的AABB包围盒模型是否相交；若不相交，认为无潜在碰撞风险发生；若相交，筛选相交区域，提起待检测外部工装列表；Step 4. According to the predicted positions and postures of the robot joints, rods, and workpieces after 5 cycles, generate a robot-workpiece AABB bounding box model, and check whether the robot-workpiece AABB bounding box model and the combined external tooling AABB bounding box model Intersect; if it does not intersect, it is considered that there is no potential collision risk; if it intersects, screen the intersecting area and bring up the list of external tooling to be tested;
生成机器人-工件的AABB包围盒模型的方法为：通过更新机器人运动状态，获取更新的机器人模型，基于基坐标系获取机器人和工件各x、y、z方向上的顶点，生成AABB包围盒模型，参见图2。The method of generating a robot-workpiece AABB bounding box model is to obtain the updated robot model by updating the robot motion state, obtain the vertices of the robot and the workpiece in the x, y, and z directions based on the base coordinate system, and generate the AABB bounding box model. See Figure 2.
在本实施例中，假定机器人-工件的AABB包围盒模型与外部工装1相交，与外部工装2、3不相交，外部工装1的列表如下：In this embodiment, it is assumed that the robot-workpiece AABB bounding box model intersects with external tooling 1, but does not intersect with external tooling 2, 3. The list of external tooling 1 is as follows:
表1 外部工装1列表Table 1 List of external tooling 1
| 序号Serial number | 名称name |
| 1.11.1 | 底座Base |
| 1.21.2 | 工作台Workbench |
| 1.31.3 | 后挡板Tailgate |
分别提取底座、工作台及后挡板的几何模型。Separately extract the geometric models of the base, the worktable and the tailgate.
步骤5，检测机器人、工件几何模型与待检测外部工装列表中的外部工装几何模型是否相交；若相交，认为发生碰撞，立即停止机器人；若不相交，认为无潜在碰撞风险发生；继续下一个控制周期；Step 5. Check whether the geometric model of the robot and the workpiece intersect with the external tool geometric model in the external tool list to be tested; if it intersects, it is considered that a collision has occurred, and the robot is stopped immediately; if it does not intersect, it is considered that there is no potential collision risk; continue to the next control cycle;
在本实施例中，检测机器人本体几何模型、工件几何模型与外部工装1中底座、工作台、后挡板的几何模型是否相交。In this embodiment, it is detected whether the geometric model of the robot body and the geometric model of the workpiece intersect with the geometric model of the base, the worktable, and the tailgate in the external tooling 1.
步骤6，下一个控制周期重复步骤2至步骤5，直到机器人完成全部动作。Step 6. Repeat Step 2 to Step 5 in the next control cycle until the robot completes all actions.
实施方式2Embodiment 2
实施方式2的方法与实施方式1的基本相同，不同之处仅在于在确认预测周期m的方法中引入补偿周期参数γ，m＝t
stop/Δt+γ；其中γ为补偿周期参数。
The method of Embodiment 2 is basically the same as that of Embodiment 1, except that the compensation period parameter γ is introduced into the method of confirming the prediction period m, m=t stop /Δt+γ; where γ is the compensation period parameter.
在本实施例中，假定补偿周期参数γ为1，周期m的计算方法即为m＝10/2+1＝6。实施方式2中的m周期数为6，相较于实施方式1，增大一个周期数。即实施方式2可在当前时刻下，预测t＝50+6*10＝110ms时刻是否有潜在碰撞发生，而实施方式1中只能预测t＝50+5*10＝100ms时刻是否有潜在碰撞发生。即实施方式2中的方法可比实施方式1更早预测碰撞发生风险，提供更充足的停止时间，以及更长的停止距离。In this embodiment, assuming that the compensation period parameter γ is 1, the calculation method of the period m is m=10/2+1=6. In Embodiment 2, the number of m cycles is 6, which is one cycle longer than in Embodiment 1. That is, Embodiment 2 can predict whether a potential collision occurs at t=50+6*10=110ms at the current time, while Embodiment 1 can only predict whether a potential collision occurs at t=50+5*10=100ms. . That is, the method in Embodiment 2 can predict the risk of collision occurrence earlier than Embodiment 1, provide more sufficient stopping time, and a longer stopping distance.
根据机器人运行工况不同，适当增大或减少γ。γ越小，碰撞检测灵敏度越小，即检测到潜在碰撞风险后机器人的停止位置与潜在碰撞外部工装之间距离越近。According to the different operating conditions of the robot, increase or decrease γ appropriately. The smaller γ, the smaller the collision detection sensitivity, that is, the shorter the distance between the robot's stopping position and the potential collision external tooling after detecting the potential collision risk.
以上说明书中未做特别说明的部分均为现有技术，或者通过现有技术既能实现。The parts that are not specifically described in the above description are all existing technologies, or can be realized by existing technologies.
Claims (7)
- 一种工业机器人防碰撞方法，其特征在于，包括顺次相接的以下步骤：An anti-collision method for industrial robots, characterized in that it includes the following steps connected in sequence:步骤一，分别建立初始状态下机器人、工件及外部工装的几何模型，及简化合并外部工装的AABB包围盒模型；Step 1: Establish the geometric model of the robot, workpiece and external tooling in the initial state, and simplify the AABB bounding box model incorporating external tooling;步骤二，机器人动作时，在第一控制周期内，根据观测的机器人运动状态预测m周期后机器人各关节、杆件、工件的位置和姿态，根据预测的机器人各关节、杆件和工件的位置和姿态刷新机器人、工件的几何模型；Step 2: When the robot is moving, in the first control cycle, predict the position and posture of each joint, rod and workpiece of the robot after m cycles according to the observed robot motion state, and based on the predicted position of each joint, rod and workpiece of the robot And posture refresh the geometric model of robot and workpiece;步骤三，检测工件与机器人本体各关节、杆件几何模型是否相交；若相交，认为工件与机器人本体将发生碰撞，立即停止机器人；若不相交，继续机器人、工件与外部工装碰撞检测；Step 3: Check whether the geometric models of the joints and rods of the workpiece and the robot body intersect; if they intersect, the workpiece and the robot body will be collided and stop the robot immediately; if they do not intersect, continue the collision detection between the robot, the workpiece and the external tooling;步骤四，根据预测m周期后机器人各关节、杆件、工件的位置和姿态，生成机器人-工件的AABB包围盒模型，检测机器人-工件的AABB包围盒模型与简化合并外部工装的AABB包围盒模型是否相交；若不相交，认为无潜在碰撞风险发生；若相交，筛选相交区域，提取待检测外部工装列表；Step 4: According to the position and posture of the robot joints, rods, and workpieces after the prediction of m cycles, generate a robot-workpiece AABB bounding box model, detect the robot-workpiece AABB bounding box model, and simplify the merged external tooling AABB bounding box model Whether it intersects; if it does not intersect, it is considered that there is no potential collision risk; if it intersects, screen the intersection area and extract the list of external tooling to be tested;步骤五，检测机器人、工件几何模型与待检测外部工装列表中的外部工装几何模型是否相交；若相交，认为发生碰撞，立即停止机器人；若不相交，认为无潜在碰撞风险发生；继续下一个控制周期；Step 5: Check whether the geometric model of the robot and the workpiece intersects the geometric model of the external tooling in the external tooling list to be tested; if it intersects, it is considered that a collision occurs, and the robot is stopped immediately; if it does not intersect, it is considered that there is no potential collision risk; continue to the next control cycle;步骤六，在下一个控制周期内重复步骤2至步骤5，直到机器人完成全部动作。Step 6: Repeat steps 2 to 5 in the next control cycle until the robot completes all actions.
- 如权利要求1所述的工业机器人防碰撞方法，其特征在于，步骤一所述机器人、工件及外部工装的几何模型为由机器人、工件、外部工装的机械属性定义的几何模型。The anti-collision method of an industrial robot according to claim 1, wherein the geometric model of the robot, the workpiece and the external tooling in step 1 is a geometric model defined by the mechanical properties of the robot, the workpiece and the external tooling.
- 如权利要求1所述的工业机器人防碰撞方法，其特征在于，步骤一所述简化合并外部工装的AABB包围盒模型的生成方法为：通过示教区域的方式将临近的外部工装的机械属性定义的几何模型合并，生成包含多个外部工装的AABB包围盒模型。The anti-collision method for industrial robots according to claim 1, wherein the method of generating the simplified AABB bounding box model incorporating external tooling in step 1 is to define the mechanical properties of nearby external tooling by means of a teaching area The geometric models are combined to generate an AABB bounding box model containing multiple external fixtures.
- 如权利要求1所述的工业机器人防碰撞方法，其特征在于，步骤二所述m周期的计算方法为：The anti-collision method for industrial robots according to claim 1, wherein the calculation method of the m period in step 2 is:获取机器人在本周期内的各关节最大速度以及最大加速度
计算各关节在最大速度下停止所需时间
即各关节最大停止时间，选取停止时间最长的关节停止时间为机器人动作停止最长时间t stop；获取控制器运算周期Δt，计算预测周期数： Get the maximum speed of each joint of the robot in this cycle
And maximum accelerationCalculate the time required for each joint to stop at the maximum speedThat is, the maximum stop time of each joint, the joint stop time with the longest stop time is selected as the maximum time for the robot to stop t stop ; the controller calculation cycle Δt is obtained, and the number of prediction cycles is calculated:
m＝t stop/Δt。 m=t stop /Δt. - 如权利要求4所述的工业机器人防碰撞方法，其特征在于，在确认预测周期m的方法中引入补偿周期参数γ，m＝t stop/Δt+γ；其中γ为补偿周期参数。 The anti-collision method for industrial robots according to claim 4, wherein the compensation period parameter γ is introduced into the method of confirming the prediction period m, m=t stop /Δt+γ; where γ is the compensation period parameter.
- 如权利要求1所述的工业机器人防碰撞方法，其特征在于，步骤二所述根据预测的机 器人各关节、杆件和工件的位置和姿态刷新机器人、工件的几何模型的方法为：The anti-collision method for industrial robots according to claim 1, characterized in that the method for refreshing the geometric model of the robot and the workpiece according to the predicted positions and postures of each joint, rod and workpiece of the robot is:⑴、从控制器中获取当前时刻机器人各关节位置θ 1,θ 2,θ 3…θ i，关节速度(1) Obtain the robot joint positions θ 1 , θ 2 , θ 3 …θ i , joint velocity from the controller at the current time
⑵、使用雅可比方程：计算各关节和工件在坐标系下合成速度V i，其中v i(t)为关节线速度，ω i(t)为t时刻关节姿态速度； ⑵、Use Jacobian equation:
Calculate the combined velocity V i of each joint and workpiece in the coordinate system, where v i (t) is the joint linear velocity, and ω i (t) is the joint posture velocity at t;
⑶、设定关节在当前时刻t做匀速直线运动，通过状态方程计算m周期后各关节、杆件和工件几何模型位置和姿态；(3) Set the joints to perform linear movement at a constant speed at the current time t, and calculate the position and posture of the geometric models of each joint, rod and workpiece after the m cycle through the state equation;P i(t+m·Δt)＝v i(t)·m·Δt+P i(t) P i (t+m·Δt)=v i (t)·m·Δt+P i (t)O i(t+m·Δt)＝ω i(t)·m·Δt+O i(t) O i (t+m·Δt)=ω i (t)·m·Δt+O i (t)其中Δt为控制器运算周期，P i(t)为控制器运算周期，Δt为控制器运算周期，P i(t)为当前时刻t下各关节、杆件和工件几何模型位置，O i(t)为当前时刻t下各关节、杆件和工件几何模型姿态，计算公式为： Where Δt is the controller calculation cycle, P i (t) is the controller calculation cycle, Δt is the controller calculation cycle, P i (t) is the position of the geometric model of each joint, rod and workpiece at the current time t, O i ( t) is the geometric model pose of each joint, rod and workpiece at the current time t, the calculation formula is:
⑷、根据机器人各关节，杆件和工件位置和姿态刷新机器人、工件几何模型。⑷ Refresh the geometric model of the robot and the workpiece according to the positions and postures of the robot joints, rods and workpieces. - 如权利要求1所述的工业机器人防碰撞方法，其特征在于，步骤四所述生成机器人-工件的AABB包围盒模型的方法为：通过更新机器人运动状态，获取更新的机器人模型，基于基坐标系获取机器人和工件各x、y、z方向上的顶点，生成AABB包围盒模型。The anti-collision method for an industrial robot according to claim 1, wherein the method for generating the robot-workpiece AABB bounding box model in step 4 is: obtaining the updated robot model by updating the robot motion state based on the base coordinate system Obtain the vertices of the robot and the workpiece in the x, y, and z directions to generate an AABB bounding box model.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811634172.9A CN109773785A (en) | 2018-12-29 | 2018-12-29 | A kind of industrial robot collision-proof method |
| CN201811634172.9 | 2018-12-29 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2020133887A1 true WO2020133887A1 (en) | 2020-07-02 |
Family
ID=66498900
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2019/087243 WO2020133887A1 (en) | 2018-12-29 | 2019-05-16 | Industrial robot collision prevention method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109773785A (en) |
| WO (1) | WO2020133887A1 (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110807831B (en) * | 2019-09-18 | 2023-03-28 | 重庆大学 | Sensor coverage area calculation method based on minimum unit collision detection |
| CN111360824B (en) * | 2020-02-27 | 2021-04-30 | 中科新松有限公司 | Double-arm self-collision detection method and computer-readable storage medium |
| CN111251335B (en) * | 2020-03-24 | 2021-06-08 | 桂林电子科技大学 | High-precision mechanical arm collision detection method based on bounding box algorithm |
| CN111267073B (en) * | 2020-03-24 | 2022-11-15 | 青岛理工大学 | Industrial robot teaching system and method based on augmented reality technology |
| CN111546376A (en) * | 2020-04-22 | 2020-08-18 | 哈尔滨工业大学 | Rapid self-collision detection method of space combination mechanical arm |
| CN113664832A (en) * | 2021-09-01 | 2021-11-19 | 上海节卡机器人科技有限公司 | Robot collision prediction method, computer storage medium and electronic device |
| CN113781846B (en) * | 2021-09-28 | 2022-12-16 | 中国人民解放军空军工程大学 | Horizontal dimension airspace conflict detection method |
| CN116638515A (en) * | 2023-05-31 | 2023-08-25 | 重庆日联科技有限公司 | Anti-collision protection method for X-ray detection system of robot in lead room |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09290392A (en) * | 1996-04-24 | 1997-11-11 | Kao Corp | Collision preventing method and device for article handling device |
| US20030225479A1 (en) * | 2002-05-30 | 2003-12-04 | El-Houssaine Waled | Method and control device for avoiding collisions between cooperating robots |
| WO2017080649A2 (en) * | 2015-11-11 | 2017-05-18 | Kuka Roboter Gmbh | Sensitive robots using sensor technology arranged before the first robot axis |
| CN107803831A (en) * | 2017-09-27 | 2018-03-16 | 杭州新松机器人自动化有限公司 | A kind of AOAAE bounding volume hierarchy (BVH)s collision checking method |
| CN108582070A (en) * | 2018-04-17 | 2018-09-28 | 上海达野智能科技有限公司 | robot collision detecting system and method, storage medium, operating system |
-
2018
- 2018-12-29 CN CN201811634172.9A patent/CN109773785A/en active Pending
-
2019
- 2019-05-16 WO PCT/CN2019/087243 patent/WO2020133887A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09290392A (en) * | 1996-04-24 | 1997-11-11 | Kao Corp | Collision preventing method and device for article handling device |
| US20030225479A1 (en) * | 2002-05-30 | 2003-12-04 | El-Houssaine Waled | Method and control device for avoiding collisions between cooperating robots |
| WO2017080649A2 (en) * | 2015-11-11 | 2017-05-18 | Kuka Roboter Gmbh | Sensitive robots using sensor technology arranged before the first robot axis |
| CN107803831A (en) * | 2017-09-27 | 2018-03-16 | 杭州新松机器人自动化有限公司 | A kind of AOAAE bounding volume hierarchy (BVH)s collision checking method |
| CN108582070A (en) * | 2018-04-17 | 2018-09-28 | 上海达野智能科技有限公司 | robot collision detecting system and method, storage medium, operating system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109773785A (en) | 2019-05-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020133887A1 (en) | Industrial robot collision prevention method | |
| JP7058126B2 (en) | Robot control device and automatic assembly system | |
| US20210053226A1 (en) | Safe operation of machinery using potential occupancy envelopes | |
| Božek | Robot path optimization for spot welding applications in automotive industry | |
| TWI769747B (en) | Method for calculating safety range of automated machine and device for controlling robot | |
| Popov et al. | Real-time external contact force estimation and localization for collaborative robot | |
| Kumar et al. | Dynamic awareness of an industrial robotic arm using time-of-flight laser-ranging sensors | |
| Kim et al. | Estimating probability of human hand intrusion for speed and separation monitoring using interference theory | |
| US11602852B2 (en) | Context-sensitive safety monitoring of collaborative work environments | |
| Uk et al. | Modeling, control, and simulation of a SCARA PRR-type robot manipulator | |
| Haddadin et al. | Holistic design and analysis for the human-friendly robotic co-worker | |
| Lv et al. | Hardware-in-the-loop simulation for evaluating communication impacts on the wireless-network-controlled robots | |
| Zhang et al. | Gesture-based human-robot interface for dual-robot with hybrid sensors | |
| US20230173682A1 (en) | Context-sensitive safety monitoring of collaborative work environments | |
| Mohammad et al. | Towards human-robot collaboration with parallel robots by kinetostatic analysis, impedance control and contact detection | |
| CN110549375A (en) | protective door anti-collision method and system for mechanical arm | |
| Johari et al. | Robotic modeling and simulation of palletizer robot using Workspace5 | |
| Yin et al. | Process simulation and optimization of ammunition filling robot based on DELMIA | |
| Zhang et al. | Trajectory tracking control of a 3-CRU translational parallel robot based on PD+ robust controller | |
| Zeng et al. | Kinematic characteristics analysis of cooperative welding robot with multiple manipulators | |
| Chen et al. | Realizing compliant insertion task based on attractive-region-in-environment | |
| Poppeová et al. | Design of ANTI-collision system for robotics | |
| Qu et al. | Dynamic Viewpoint-Based Obstacle Avoidance of Robotic Arms in Complex Environments | |
| Chen et al. | Contact-aware shaping and maintenance of deformable linear objects with fixtures | |
| Kramar et al. | The predicting collision with external obstacles of dual-arm multi-link robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19903880 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 19903880 Country of ref document: EP Kind code of ref document: A1 |