JP4952959B2 - Transport system and robot control method - Google Patents

Description

  The present invention relates to a technique for transporting an object by a robot, and more particularly, to a technique for controlling the moving direction of a robot that pushes and conveys an object.

  Object transportation is one of the most important tasks of mobile robots. If the object transportation between arbitrary places is automated by the robot, it is very beneficial because it saves the user time and labor. Various types of transfer robots have been proposed. Here, attention is paid to a type in which an object placed on a plane is pushed by a self-propelled robot and moved to a target position. This is because most transport tasks can be handled by this method (push transport), and push transport can be applied to transport objects in various environments such as factories, offices, stores, and homes. Non-Patent Document 1 discloses a home robot that performs push conveyance.

S. Zhao, K. Nakamura, K. Ishii, and T. Igarashi, "Magic Cards: A Paper Tag Interface for Implicit Robot Control", Proceedings of the ACM Conference on Human Factors in Computing Systems, CHI2009, pp.173-182 , Boston, USA, April, 2009.

  When an object is pushed by a robot, the object does not always move to a planned position. This is because various factors such as friction between the object and the surface on which the object is placed, interference between the outer shape of the object and the robot, displacement of the center of gravity of the object and the pushing position can affect the movement of the object. Therefore, in push transport by a self-propelled robot, it is necessary to sequentially control the movement (position and orientation) of the robot so that the object is pushed toward the target position while the system constantly monitors the position of the object. .

As shown in FIG. 9A, the simplest algorithm for pushing and conveying is that the robot R is first moved to the pushing position behind the object O, and then the robot R is moved toward the target position G. Things are known (this method is called binary mode method). When the object O deviates from the path, the robot R stops the pushing operation and moves to return to the pushing position. However, this binary mode method is not very efficient because a large overhead is required for the return operation. As another algorithm, as shown in FIG. 9B, a method of moving the robot R toward a position slightly behind the center of the object O can be considered (this method is called a tail-following method). Although this method is more efficient than the binary mode method, an error may occur when, for example, the robot is closer to the target position than the object. In addition, since each method is an algorithm that depends on the size of the object and the robot, the program is changed according to the environment (object size, robot size, etc.) to which the system is applied, or various parameters depending on the size are set. There is also a problem that it must be changed or very troublesome.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an algorithm capable of efficiently and stably performing push conveyance of a robot. A further object of the present invention is to provide a highly versatile push transport algorithm that is simple and does not depend on the size of an object or a robot.

  In order to achieve the above object, in the present invention, an operation in which the robot pushes the object (pushing operation) and an operation in which the robot moves around the object and moves toward the pushing position on the back side of the object (circular operation) are appropriately performed. An algorithm that determines the moving direction of the robot by combining with weights was adopted.

Specifically, the transport system according to the present invention is a transport system that transports the object to a predetermined target position by pushing the object with a robot, the self-propelled robot for pushing the object, the object, Position detection means for detecting the current position of the robot, and control means for controlling the movement direction of the robot based on the detection result of the position detection means, wherein the control means is configured such that the robot moves the position of the object. A first vector in a direction toward the first direction and a second vector orthogonal to the first vector and in a direction in which the robot leaves the target position, the current positional relationship between the object, the robot, and the target position. by synthesizing in accordance with the weight, which determines the movement direction of the robot, to the robot from the vector and the object to the target position from the object When the angle formed by the vector is θ, the first vector is (−cos θ, −sin θ), the second vector is (−sin θ, cos θ), and the first vector is Is a cos θ, and the weight of the second vector is sin θ .

  By moving the robot according to the combined vector of the first vector and the second vector, it is possible to execute both the pushing operation and the rounding operation at the same time, thereby realizing efficient push conveyance. At this time, further efficiency improvement can be expected by appropriately adjusting the weights of the pushing operation and the orbiting operation according to the positional relationship between the robot, the object, and the target position. Further, it is only necessary to perform vector calculation from the three values of the current position and the target position of the robot and the object, which is an extremely simple algorithm. Furthermore, since the first vector, the second vector, and their weights do not depend on the size of the robot or the object, the algorithm of the present invention can be applied universally to any size object or robot.

Further , the weight changes continuously according to θ, and when θ is 180 degrees, the weight of the first vector takes the maximum value 1 and the weight of the second vector takes zero. Further, as θ becomes smaller than 180 degrees, the weight of the second vector gradually increases, and accordingly, the weight of the first vector gradually decreases. Therefore, as the robot moves away from just behind the object, the orbiting motion component Increases and the return to the optimal pushing position (θ = 180 degrees) is promoted. When θ is 90 degrees, that is, when the robot is directly beside the object, the weight of the first vector becomes zero and the weight of the second vector becomes maximum, so that the object can be moved away from the target position. And a quick return to the optimum pushing position can be expected.

  The control means preferably sets the weight of the first vector to zero when θ is 90 degrees or less.

  When θ is 90 degrees or less, if the weight of the first vector is larger than zero, the object is moved away from the target position, and if the weight of the first vector is smaller than zero, the robot itself is moved away from the object. Both are inefficient. On the other hand, if θ is 90 degrees or less, the weight of the first vector is fixed to zero and only the rounding operation is performed, so that it is possible to increase the efficiency of push conveyance.

  Preferably, the control unit increases the weight of the second vector when the object approaches the target position. Typically, the control means may multiply the weight of the second vector by a coefficient a (a ≧ 1) that increases as the distance between the object and the target position becomes shorter.

  As a result, the orbiting motion component increases in the vicinity of the target position, and the pushing motion component becomes relatively small, so that the pushing motion is performed carefully (a little by little). Therefore, it is possible to prevent the object from deviating from the path or overshooting the target position, and as a result, the convergence to the target position is accelerated.

  The transport system further includes a second robot that is self-propelled, the position detecting unit further detects a current position of the second robot, the control unit is configured to detect the robot by the second robot. By considering the object to be conveyed and controlling the movement direction of the second robot using the same algorithm as in the movement direction of the robot, the robot and the second robot are connected in series, and the object It is also preferable to press.

  As a result, a heavy object that is difficult to be transported by one robot can be transported by the cooperation of a plurality of robots. In addition, since the algorithm for pressing the object by the first robot and the algorithm for pressing the first robot by the second robot can be made the same, it is easy to create and implement the program.

  The present invention can be understood as a transport system or a robot control system having at least a part of the above means. The present invention can also be understood as a robot control method including at least a part of the above processing, a program for causing a computer to execute each step of the method, or a recording medium on which the program is recorded. Each of the above means and processes can be combined with each other as much as possible to constitute the present invention.

For example, a robot control method according to one aspect of the present invention is a robot control method in a transport system that transports an object to a predetermined target position by pushing the object with a self-propelled robot, the object and the robot Detecting a current position of the robot, and a control step of controlling a moving direction of the robot based on a detection result of the position detecting step. In the control step, the robot moves toward the position of the object. And a second vector orthogonal to the first vector and in a direction in which the robot moves away from the target position, a weight corresponding to the current positional relationship between the object, the robot, and the target position by combining in, to determine the direction of movement of the robot, from said vector and the object to the target position from the object b And the second vector is (−sin θ, cos θ), and the second vector is (−sin θ, cos θ). The robot control method is characterized in that the weight of the first vector is −cos θ and the weight of the second vector is sin θ .

  ADVANTAGE OF THE INVENTION According to this invention, the algorithm which can perform the pushing conveyance of a robot efficiently and stably can be provided. Further, according to the present invention, it is possible to provide a highly versatile push transport algorithm that is simple and does not depend on the size of an object or a robot.

It is a figure which shows the hardware constitutions of a conveyance system, (a) is a side view, (b) is a top view. It is a block diagram which shows the function structure of a conveyance system. It is a figure which shows an example of the positional relationship of a robot, an object, and a target position. (A) is a figure which shows the change of the moving direction of a robot when the positional relationship of a robot and an object is changed, (b) is a figure which shows a vector field (cos2 (theta), sin2 (theta)). It is a flowchart of a conveyance system. (A) is a figure which shows the interference of a linear bumper and an object, (b) is a figure which shows the conveyance operation by a circular bumper. It is a figure for demonstrating the definition of the coefficient a of 3rd Embodiment. It is a figure which shows the mode of the serial conveyance in 4th Embodiment. (A) is a figure which shows binary mode method, (b) is a figure which shows tail-following method.

  Exemplary embodiments of the present invention will be described in detail below with reference to the drawings.

<First Embodiment>
(System configuration)
FIG. 1 shows a hardware configuration of a transport system according to an embodiment of the present invention. (A) is a side view, (b) is a plan view of the stage S as viewed from above. This system is a transport system for transporting an object O placed on a stage S to a target position G by a robot R, and includes a self-propelled robot R, an imaging device 2 and a computer 3. Is done. The robot R and the imaging device 2 are connected to the computer 3 by wireless or wired, respectively.

  The robot R is a self-propelled robot including a moving mechanism including wheels, a drive motor, and the like, and a microcontroller that controls the moving mechanism based on a control command from the computer 3. The robot R of the present embodiment has a substantially disk-shaped outer shape, and a circular bumper is attached to the outer periphery of the robot R in order to prevent the object from being caught or damaged. . The movement mechanism of the robot R can perform at least “advance”, “clockwise”, “counterclockwise”, and “stop” operations. These operations are programmed in advance in the microcontroller, and the operation of the robot R can be controlled by sending a control command corresponding to each operation from the computer 3.

  The imaging device 2 is a small camera having an imaging element such as a CCD or a CMOS, and is installed above the planar stage S so that the stage S is held at an angle of view. In the present embodiment, the imaging device 2 is fixed so that the optical axes of the stage S and the imaging device 2 are substantially orthogonal to each other. However, the present invention is not limited to this configuration, and the imaging device 2 is installed so that the stage S is viewed obliquely. Alternatively, the angle of view and the direction of the optical axis of the imaging device 2 may be variable. The image acquired by the imaging device 2 is used for position detection of the robot R and the object O (details will be described later).

FIG. 2 is a block diagram illustrating a functional configuration of the transport system. The system includes an outline, a position detection unit 30, a movement direction calculation unit 31, and a command transmission unit 32 as its functions.
These functions are realized by the CPU of the computer 3 loading and executing a program stored in the storage device. Note that a general-purpose personal computer including a CPU, a memory, an auxiliary storage device, an input device, a display device, and the like can be used as the computer 3. However, it is also preferable to use a dedicated computer or an on-board microcomputer mounted on a robot. It is also possible to configure part or all of the program with a dedicated IC.

  The position detection unit 30 is position detection means for detecting the current positions of the robot R and the object O. Here, the position of the robot R refers to the position of the representative point set on the robot R. The position of the object O refers to the position of the representative point set on the object O. Each representative point can be set arbitrarily, but considering the stability of movement of the robot R and the object O, it is preferable to set the rotation center of the robot R and the center of mass (center of gravity) of the object O as representative points. In the present embodiment, as shown in FIG. 1, markers Mr and Mo on which predetermined figures are drawn are attached to the robot R and the object O, respectively. The marker attachment position is determined based on the representative point. The position detection unit 30 detects the positions and orientations of the robot R and the object O by recognizing the markers Mr and Mo from the image captured from the imaging device 2. In addition, since a well-known figure recognition method can be utilized for the recognition of a marker, detailed description is abbreviate | omitted here.

  The movement direction calculation unit 31 is a function that calculates the optimal movement direction of the robot R based on the detection result of the position detection unit 30. In addition, the command transmission unit 32 determines the control command of the robot R (here, “forward”, “clockwise”, “counterclockwise”, “counterclockwise”) from the current direction of the robot R and the movement direction calculated by the movement direction calculation unit 31. This is a function of generating a combination of “stop” and transmitting it to the microcontroller of the robot R. In the present embodiment, the movement direction calculation unit 31 and the command transmission unit 32 constitute a control unit of the present invention. Below, after explaining in detail the push conveyance algorithm which the movement direction calculation part 31 of this embodiment employ | adopts, the processing flow of the whole system is demonstrated.

(Push transport algorithm)
The push transport algorithm is a calculation method for calculating a suitable moving direction of the robot R for causing the object O to reach the target position G when a positional relationship as shown in FIG. The input of this algorithm is the target movement direction Do of the object O (the direction of the vector from the object O to the target position G) and the positional relationship between the robot R and the object O. The output of this algorithm is the movement direction of the robot R. Dr. Here, it is assumed that the positions of the robot R and the object O are acquired from the position detection unit 30 each time, and the target position G is known in advance. In order to physically drive the robot R, it is necessary to adjust parameters such as motor torque and timing, but the calculation of the moving direction Dr of the robot R is independent of these low-level drive control problems. Can be handled.

  When acquiring the positions of the robot R and the object O, the system (movement direction calculation unit 31) first defines a local coordinate system. The position of the object O is taken as the origin of the local coordinate system, and the x-axis is taken in parallel with the target movement direction Do of the object O. Here, assuming that the angle formed by the position vector of the robot R (the vector from the object O to the robot R) and the x axis is θ and the distance between the robot R and the object O is 1, the coordinates of the robot R are (cos θ, sin θ ).

As a basic operation of the robot R, an operation in which the robot R pushes the object O (pushing operation) and an operation in which the robot R moves around the object O and turns around the object O (circular operation) are considered. In the coordinate system, the moving direction of the pushing operation is given by a vector in a direction in which the robot R moves toward the position of the object O, specifically, a vector V1 = (− cos θ, −sin θ). The moving direction of the orbiting motion is given by a vector in a direction in which the robot R moves away from the target position G among vectors orthogonal to the vector V1, specifically, vector V2 = (− sin θ, cos θ). In contrast to the conventional binary mode method in which the pushing operation and the circular motion are executed separately, this algorithm circulates while pushing by combining the two vectors V1 and V2 with appropriate weights (or , Press while rotating) to realize the movement of the robot R.

When the robot R is behind the object O as viewed from the target position G (that is, when −cos θ is large), the pushing operation by the robot R should be dominant. On the other hand, as the robot R moves away from the position immediately behind the object O (that is, as sin θ is larger), the orbiting operation of the robot R should be dominant. Therefore, in the present algorithm, −cos θ and sin θ are selected as the weights of the vectors V1 and V2. Then, the moving direction Dr of the robot R is given by (cos2θ, sin2θ) as shown in the following equation.

  FIG. 4A shows a change in the moving direction Dr of the robot R when the positional relationship between the robot R and the object O is changed. FIG. 4B shows a vector field (cos 2θ, sin 2θ). From these figures, it can be seen that the pushing operation becomes dominant as the robot R moves to the rear side of the object O, and that the moving direction Dr continuously changes according to the positional relationship between the robot R and the object O.

  What is more interesting is that the vector field in FIG. 4 (b) resembles a dipole in physics (thus, this algorithm is also called the dipole method). A dipole in physics is, for example, an electric field that is observed when positive and negative charges exist apart. Geometrically, as shown in FIG. 4 (b), a dipole is drawn with a large number of circles in contact with the straight line at one point on the straight line.

(Processing flow)
With reference to the flowchart of FIG. 5, the process flow of the present system will be described. It is assumed that the robot R and the object O are arranged on the stage in advance, and position detection markers Mr and Mo are attached to them. Further, it is assumed that the target position G, which is a transfer goal, is previously taught to the system.

  In step S <b> 10, when the position detection unit 30 captures an image from the imaging device 2, the position detection unit 30 identifies the current positions and orientations of the robot R and the object O by recognizing the markers Mr and Mo. Next, the movement direction calculation unit 31 calculates the target movement direction Do of the object O from the target position G and the current position of the object O obtained in step S10 (step S12). Then, the movement direction calculation unit 31 calculates the movement direction Dr of the robot R using the push conveyance algorithm (dipole method) described above (step S13).

  Next, the command transmission unit 32 controls the movement of the robot R according to the movement direction Dr calculated in step S13. Specifically, the command transmission unit 32 compares the direction of the robot R obtained in step S10 and the movement direction Dr (step S14), and if the difference is smaller than a predetermined threshold Th2, the command transmission unit 32 determines the robot R as the predetermined direction. Move forward by a distance (step S15). On the other hand, if the difference is greater than or equal to the threshold Th2, the command transmission unit 32 sends a clockwise or counterclockwise command to adjust the orientation of the robot R (step S16).

  The system repeats the above processing, and when the distance between the object O and the target position G becomes smaller than the predetermined threshold Th1, the system determines that the conveyance is complete and ends the processing (step S11).

(Advantages of this embodiment)
According to the configuration (algorithm) of the present embodiment described above, the following advantages are obtained.
Since the robot R is moved according to the combined vector of the vectors V1 and V2, both the pushing operation and the rounding operation can be performed at the same time, so that an efficient push conveyance is realized as compared with the conventional algorithm. In addition, by selecting the weights of the vectors V1 and V2 to be −cos θ and sin θ, respectively, the moving direction of the robot smoothly changes according to θ (the positional relationship between the robot, the object, and the target position). Operation is stable. Further, when θ is 180 degrees, the weight of the vector V1 (push) takes the maximum value 1, and the weight of the vector V2 (lap) takes zero, so that only the push operation is performed when the robot is directly behind the object. As a result, the efficiency of push conveyance can be maximized. Further, as θ becomes smaller than 180 degrees, the weight of the vector V2 (circulation) gradually increases, and the weight of the vector V1 (push) gradually decreases accordingly. The movement component becomes large, and the return to the optimal pushing position (θ = 180 degrees) is promoted. When θ is 90 degrees, that is, when the robot is directly beside the object, the weight of the vector V1 becomes zero and the weight of the vector V2 becomes the maximum, so that the object is not moved away from the target position, and A quick return to the optimal pushing position can be expected.

  In addition, this algorithm only needs to perform vector calculation from the three values of the current position and target position of the robot and the object, and is extremely simple and requires a very small amount of calculation. Further, the moving direction Dr of the robot is given by (cos 2θ, sin 2θ), and does not depend on the size of the robot or the object, the distance between the robot and the object, or the like. Therefore, according to the present algorithm, the same program code can be used for any size object or robot, and complicated operations such as adjustment of size-dependent parameters are not required. Thereby, the labor for practical use can be greatly reduced.

  In addition, since the robot according to the present embodiment has a circular bumper, it can be more efficiently pushed and conveyed than a linear bumper. This point will be described with reference to FIG. When the robot R is directly behind the object O, any bumper shape can be transported in the same manner. However, if the position of the robot R is shifted from just behind the object O, when the robot R turns for a circular motion, the linear bumper may interfere with the object O and move the object O away from the target position. (See FIG. 6 (a)). In order to prevent bumper interference, it is necessary to revolve the robot R and once away from the object O, and then perform a revolving operation. However, such an operation is not preferable because it causes a great overhead. On the other hand, as shown in FIG. 6B, in the case of a circular bumper, there is no interference between the bumper and the object O when the robot R turns, and when the robot R finishes turning and moves forward, O is pushed toward the target position. Therefore, a circular bumper is more suitable for the push conveyance algorithm of this embodiment.

Second Embodiment
When the robot R is in front of the object O, according to the algorithm of the first embodiment, the robot R moves in a direction away from the object O and draws a large arc and goes around the object O in an inefficient trajectory. (See FIG. 4B). For example, the above problem occurs when the robot R pushes the object O too far before the target position G and the object O passes the target position G.

In order to solve such a problem, in the algorithm of the second embodiment, when the angle θ is 90 degrees or less, the weight of the vector V1 in the pushing operation direction is set to zero as shown in the following equation. this is,
In other words, when the moving direction Dr of the robot obtained by the algorithm of the first embodiment is not directed toward the object O, the radial component of Dr is set to zero.

In addition to being easy to understand and easy to implement, this method has the following advantages. First, a vector field (−sin ² θ, sin θ cos θ) at 0 degree <θ ≦ 90 degrees and a dipole field (cos 2θ, sin 2θ) at 90 degrees <θ ≦ 180 degrees are the boundaries (θ = 90 degrees). Connect smoothly. This has the effect of stabilizing the operation of the robot R near the boundary. Secondly, the algorithm of this embodiment is excellent in versatility because it does not depend on the size of the robot or the object as in the first embodiment.

<Third Embodiment>
In the algorithms of the first and second embodiments, it may take time for the object O to converge to the target position G. In the worst case, the robot R and the object O may continue to rotate around the target position G indefinitely. Such a phenomenon occurs because, in the dipole method, the robot R moves around behind the object O and simultaneously pushes and moves the object O.

In order to solve such a problem, in the algorithm of the third embodiment, when the object O approaches the target position G, the weight of the vector V2 of the orbital motion component is increased. The moving direction Dr of the robot R of the present embodiment is expressed by the following formula. Here, the coefficient a (a ≧ 1) is a parameter determined according to the distance between the object O and the target position G.

  As the coefficient a increases, the orbiting motion component increases and the push motion component relatively decreases, so that the push motion is performed carefully (a little by little). Therefore, if the coefficient a is increased in the vicinity of the target position G, it is possible to prevent the object O from moving out of the transport path or overshooting the target position G, and as a result, the convergence to the target position G is accelerated. . As is clear from the above equation, the algorithm of this embodiment also has an advantage that it does not depend on the size of the robot R or the object O.

In the present embodiment, the coefficient a is defined as a = | β | / | α |. Here, α is an angle between a vector from the robot R to the object O and a vector from the robot R to the target position G, and β is a vector from the object O to the robot R and a vector from the target position G to the object O. (See FIG. 7). This coefficient a has the following characteristics. First, since the coefficient a is inversely proportional to the angle α, the coefficient a tends to increase as the object O approaches the target position G. Second, by multiplying by the angle β, the coefficient a is prevented from becoming infinite when the robot R, the object O, and the target position G are arranged in a substantially straight line. Thirdly, the parameters are independent of the size of the robot or the object because they are defined by the angles α and β.

<Fourth embodiment>
FIG. 8 shows a state of object conveyance in the fourth embodiment. In this embodiment, one robot is pushed and conveyed using two robots.

  The system of the present embodiment specifies the positions and orientations of the first robot (first robot) R1, the second robot (following robot) R2, and the object O from the image obtained by the imaging device 2. And a system determines the moving direction of robot R1 using the same algorithm as 1st-3rd embodiment. Furthermore, the system applies the same algorithm to the calculation of the moving direction of the robot R2. At this time, the robot R1, not the object O, is set as the object to be transported by the robot R2 (specifically, the positions of the robot R2 and the robot R1 are given as input of the algorithm). R2 is connected in series and pushes the object O.

  As a result, a heavy object that is difficult to be transported by one robot can be transported by the cooperation of a plurality of robots. In addition, since the algorithm for pressing the object by the first robot and the algorithm for pressing the first robot by the subsequent robot can be made the same, it is easy to create and implement the program. In this embodiment, serial transfer by two robots is illustrated, but it goes without saying that serial transfer by three or more robots is also possible.

<Other variations>
The above embodiment is merely an example of the present invention. The scope of the present invention is not limited to the above embodiment, and various modifications can be made within the scope of the technical idea.

  For example, the shape of the robot may not be a disk shape. The bumper may be provided only on a part of the outer periphery of the robot, or may be a flat shape instead of a circle. Moreover, the moving mechanism of the robot is not limited to wheels, and may be a caterpillar, a levitating device, a walking device, or the like. Any configuration of the robot can be adopted as long as the object can be pushed by moving two-dimensionally on the plane and coming into contact with the object. Further, the outer shape of the object to be transported does not need to be circular, and any object having an outer shape that can contact the bumper of the robot, such as an ellipse, a rectangle, or a polygon, can be transported.

  In the above embodiment, the position of an object or the like is detected from the image obtained by the imaging device 2, but the position detection means is not limited to this configuration. For example, a sensor may be built in the stage, and two-dimensional coordinates on the stage of the object or robot may be detected by the sensor. Alternatively, a position sensor may be built in the object or the robot itself.

  The definition of the coefficient a used in the third embodiment is not limited to that described above. Further, it is not necessary to continuously change the value of the coefficient a in accordance with the distance between the object and the target position, and a configuration in which several values are switched in stages may be used. For example, when the distance between the object O and the target position G is equal to or greater than a predetermined threshold (when the distance is not near the target position), a = 1, and when the distance is smaller than the predetermined threshold (when the distance is close to the target position), a = 2. It is also possible to use the method.

  R, R1, R2: Robot, O: Object, G: Target position, S: Stage, 2: Imaging device, 3: Computer, 30: Position detection unit, 31: Movement direction calculation unit, 32: Command transmission unit

Claims (6)

A transport system that pushes an object by a robot and transports the object to a predetermined target position,
A self-propelled robot to push objects,
Position detecting means for detecting a current position of the object and the robot;
Control means for controlling the movement direction of the robot based on the detection result of the position detection means,
The control means includes a first vector in a direction in which the robot moves toward the position of the object, and a second vector in a direction orthogonal to the first vector and in a direction in which the robot moves away from the target position. The moving direction of the robot is determined by synthesizing with a weight according to the current positional relationship between the robot and the target position ,
When the angle formed by the vector from the object to the target position and the vector from the object to the robot is θ,
The first vector is (−cos θ, −sin θ);
The second vector is (−sin θ, cos θ);
The weight of the first vector is −cos θ,
The transport system according to claim 1, wherein a weight of the second vector is sin θ . The transport system according to claim 1 , wherein the control unit sets the weight of the first vector to zero when θ is 90 degrees or less. Wherein, the transport system according to claim 1 or 2, wherein the object is characterized by increasing the weight of the second vector when approaching the target position. A self-propelled second robot,
The position detecting means further detects a current position of the second robot;
The control means regards the robot as an object carried by the second robot, and controls the moving direction of the second robot using the same algorithm as in the moving direction of the robot.
Conveying system according to any one of claims 1 to 3, and the robot and the second robot, characterized in that pressing the said object connected in series. A method for controlling a robot in a transport system that transports the object to a predetermined target position by pushing the object with a self-propelled robot,
Detecting the current position of the object and the robot;
A control step for controlling a moving direction of the robot based on a detection result of the position detection step,
In the control step, a first vector in a direction in which the robot moves toward the position of the object, and a second vector in a direction orthogonal to the first vector and in a direction in which the robot moves away from the target position are By combining with a weight according to the current positional relationship between the robot and the target position, determine the movement direction of the robot ,
When the angle formed by the vector from the object to the target position and the vector from the object to the robot is θ,
The first vector is (−cos θ, −sin θ);
The second vector is (−sin θ, cos θ);
The weight of the first vector is −cos θ,
A robot control method, wherein the weight of the second vector is sin θ . The program for making a computer perform each step of the control method of the robot of Claim 5 .

Images