forklift
Patent Information
- Application Number
- JP2025059217
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-03-31
- Publication Date
- 2026-10-14
AI Technical Summary
【0008】 本発明によれば、載置面と荷とのずれ角度を算出することができる。
Smart Images

Figure 2026166341000001_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to a forklift.
Background Art
[0002] A forklift disclosed in Patent Document 1 places a load on the cargo bed of a truck. The forklift includes forks, a laser scanner, and a control device. The forklift travels to the vicinity of the truck while holding the load on the forks. Then, after moving along the front-rear direction of the truck, the forklift turns to directly face the cargo bed. The control device calculates the position of the tip of the truck's tailgate from the detection result of the laser scanner. The control device advances the forklift to a position a predetermined distance in front of the position of the tip of the tailgate. Then, the control device operates the forks to place the load on the cargo bed.
Prior Art Documents
Patent Documents
[0003]
Patent Document 1
Summary of the Invention
Problems to be Solved by the Invention
[0004] When placing a load on the cargo bed, if the forks enter obliquely relative to the edge of the cargo bed, the load is placed obliquely with respect to the front-rear direction of the truck. A forklift may be required to calculate the misalignment angle between the upper surface of the cargo bed and the load.
Means for Solving the Problems
[0005] A forklift that solves the above problems is a forklift for loading cargo onto a truck equipped with side panels on the sides of the cargo bed, comprising: two forks spaced apart from each other in the left-right direction; a first distance meter that moves in conjunction with the two forks and detects the distance to an object below the two forks; a second distance meter that moves in conjunction with the two forks and detects the distance to an object below the two forks, and is positioned apart from the first distance meter in the left-right direction; and a control device, wherein the control device advances the two forks so that they pass over the side panels, detects a first position point in time when the first distance meter is above the upper end of the side panels based on the change in the detected value of the first distance meter as the two forks advance, detects a second position point in time when the second distance meter is above the upper end of the side panels based on the change in the detected value of the second distance meter as the two forks advance, and calculates the displacement angle between the loading surface, which is the top surface of the cargo bed, and the cargo based on the distance the two forks have moved forward between the first position point and the second position point.
[0006] The control device can calculate the displacement angle between the mounting surface and the load based on the distance the fork has traveled forward between the first position and the second position. With respect to the above-mentioned forklift, the control device generates a path to the loading position set on the aforementioned loading surface, moves the forklift along the path and places the load on the aforementioned loading surface, and places multiple loads on the aforementioned loading surface from the same direction. If arranging multiple loads on the aforementioned loading surface constitutes one cycle, the control device may calculate the displacement angle when placing the load on the aforementioned loading surface for the first time in the cycle, and correct the path when placing the load on the aforementioned loading surface for the second time or later in the cycle based on the displacement angle previously calculated within the same cycle.
[0007] With respect to the above-mentioned forklift, the control device may rotate the forklift in place by the amount of the displacement angle before placing the load on it. [Effects of the Invention]
[0008] According to the present invention, the displacement angle between the mounting surface and the load can be calculated. [Brief explanation of the drawing]
[0009] [Figure 1] Figure 1 is a side view of a forklift. [Figure 2] Figure 2 is a perspective view showing a magnified portion of a forklift. [Figure 3] Figure 3 is a perspective view showing a magnified portion of the cargo handling equipment. [Figure 4] Figure 4 is a schematic diagram of a forklift. [Figure 5] Figure 5 is a plan view of the truck. [Figure 6] Figure 6 is a magnified view of the stepped section and the side panel. [Figure 7] Figure 7 is a flowchart showing the control when placing the pallet for the first time. [Figure 8] Figure 8 is a flowchart showing angular displacement detection. [Figure 9] Figure 9 shows the deviation in the fork's entry angle relative to the mounting surface. [Figure 10] Figure 10 shows the relationship between the rangefinder's detected value and time. [Figure 11] Figure 11 is a flowchart showing the control process when placing a pallet for the second time or later. [Figure 12] Figure 12 shows the relationship between the moving average difference of the rangefinder's detected values and time. [Modes for carrying out the invention]
[0010] An embodiment of a forklift will be described below. In the following description, front, back, left, right, up, and down refer to the front, back, left, right, up, and down relative to the forklift, unless otherwise specified. <Forklift> As shown in Figure 1 and Figure 2, the forklift 10 is a reach-type forklift. The forklift 10 automatically performs cargo handling operations.
[0011] The forklift 10 includes a vehicle body 11, two reach legs 12, two front wheels 13, a rear wheel 14, a head guard 15, and a cargo handling device 21. The two reach legs 12 are arranged spaced apart from each other in the left-right direction. The reach legs 12 extend forward from the vehicle body 11. One front wheel 13 is provided on each reach leg 12. The rear wheel 14 is provided on the vehicle body 11. The rear wheel 14 is a steering wheel. The rear wheel 14 is a drive wheel. The head guard 15 is disposed above the vehicle body 11.
[0012] The cargo handling device 21 is provided in front of the vehicle body 11. The cargo handling device 21 includes a mast 22, a lift bracket 25, two forks 26, a lift cylinder 31, and a reach cylinder 32.
[0013] The mast 22 is a multi-stage mast. The mast 22 includes an outer mast 23 and an inner mast 24. The inner mast 24 is provided so as to be capable of moving up and down relative to the outer mast 23.
[0014] The lift bracket 25 is provided so as to be capable of moving up and down relative to the inner mast 24. The two forks 26 are fixed to the lift bracket 25. As shown in Figure 3, the two forks 26 are arranged spaced apart from each other in the left-right direction. One of the two forks 26 is designated as the first fork 26L, and the other is designated as the second fork 26R. The first fork 26L is, for example, the left fork. The second fork 26R is, for example, the right fork. The fork 26 includes a support portion 27. The support portion 27 extends in the front-rear direction. The support portion 27 has a plate shape.
[0015] As shown in Figure 1, a pallet P is loaded on the fork 26. The pallet P is an example of a load. The pallet P is provided with a hole H. In a state where the support portion 27 is inserted into the hole H, the upper surface of the support portion 27 supports the pallet P, whereby the pallet P is loaded on the fork 26.
[0016] The lift cylinder 31 is a hydraulic cylinder. The lift bracket 25 moves up and down by supply and discharge of hydraulic oil to and from the lift cylinder 31. The fork 26 moves up and down together with the lift bracket 25.
[0017] The reach cylinder 32 is a hydraulic cylinder. The mast 22 moves in the front-rear direction by supply and discharge of hydraulic oil to and from the reach cylinder 32. The fork 26 moves in the front-rear direction together with the mast 22.
[0018] As shown in Figure 4, the forklift 10 includes a drive mechanism 51 and a hydraulic mechanism 52. The drive mechanism 51 is a member for causing the forklift 10 to perform a traveling operation. The drive mechanism 51 includes a drive source for driving the rear wheels 14 and a steering mechanism for steering the rear wheels 14. If the forklift 10 uses a motor as the drive source, the drive mechanism 51 includes a motor driver. If the forklift 10 uses an engine as the drive source, the drive mechanism 51 includes a fuel injection device.
[0019] The hydraulic mechanism 52 is a member for controlling supply and discharge of hydraulic oil to and from hydraulic equipment. The hydraulic equipment includes the lift cylinder 31 and the reach cylinder 32. The hydraulic mechanism 52 includes a pump that discharges hydraulic oil, and a control valve that controls supply and discharge of hydraulic oil to and from the hydraulic equipment. The cargo handling device 21 operates by the supply and discharge of hydraulic oil by the hydraulic mechanism 52.
[0020] The forklift 10 is equipped with an external sensor 61. The external sensor 61 detects the position of an object in a three-dimensional coordinate system. The external sensor 61 is, for example, a millimeter-wave radar, a stereo camera, a ToF (Time of Flight) camera, or a LIDAR (Laser Imaging Detection and Ranging). In this embodiment, a LIDAR is used as the external sensor 61. The external sensor 61 emits lasers in multiple directions. The external sensor 61 derives the distance to a point by receiving the reflected light reflected from the point where the laser strikes using a photodetector. The point where the laser strikes represents a part of the surface of an object. The position of the point can be expressed in polar coordinates. The coordinates of the point in polar coordinates are converted to coordinates in Cartesian coordinates. The external sensor 61 derives the coordinates of the point in the sensor coordinate system. The sensor coordinate system is a three-axis Cartesian coordinate system with the external sensor 61 as the origin. The external sensor 61 is often placed at a high location so that the laser is not obstructed. For example, the external sensor 61 is located on the head guard 15.
[0021] The forklift 10 includes a first distance meter 62 and a second distance meter 63. The distance meters 62 and 63 are, for example, laser distance meters. The distance meters 62 and 63 detect the distance to the point where the laser strikes by receiving reflected light from the point where the laser strikes. If the distance meters 62 and 63 cannot receive reflected light despite irradiating with a laser, the measured value is set to infinity. An infinity measured value means that there is no point within the measurable distance of the distance meters 62 and 63 where the laser strikes. In this embodiment, the distance meters 62 and 63 are one-dimensional distance meters capable of detecting distance in only one direction.
[0022] The distance meters 62 and 63 are attached to the cargo handling device 21. More specifically, the distance meters 62 and 63 are attached to a location where they move in conjunction with the forks 26. For example, the distance meters 62 and 63 are attached to the forks 26. As a result, the distance meters 62 and 63 move in conjunction with the two forks 26.
[0023] The second distance meter 63 is positioned separately from the first distance meter 62 in the left-right direction. The first distance meter 62 is located on the support portion 27 of the first fork 26L. The second distance meter 63 is located on the support portion 27 of the second fork 26R. The distance from the tip of the support portion 27 of the first fork 26L to the first distance meter 62 is the same as the distance from the tip of the support portion 27 of the second fork 26R to the second distance meter 63. In other words, the first distance meter 62 and the second distance meter 63 are located at the same position in the front-rear direction.
[0024] The two distance meters 62 and 63 are arranged such that the lasers emitted from the first distance meter 62 and the lasers emitted from the second distance meter 63 are directed downwards. Therefore, the distance meters 62 and 63 detect the distance to objects below the fork 26. The distance meters 62 and 63 are arranged so that the lasers are not obstructed by the pallet P even when the support part 27 is supporting the pallet P. For example, the distance meters 62 and 63 may be installed in a part of the support part 27 that is not inserted into the hole H. The laser emitted from the first distance meter 62 and the laser emitted from the second distance meter 63 are parallel to each other. In this embodiment, "parallel" and "identical" mean that errors caused by dimensional tolerances of each component and mounting of each component are tolerant.
[0025] The forklift 10 is equipped with a reach sensor 64. The reach sensor 64 detects the reach. The reach is the amount of forward movement of the forks 26, with the state where the forks 26 are closest to the vehicle body 11 being defined as 0. In other words, the reach is the distance the forks 26 have moved forward.
[0026] The forklift 10 is equipped with a camera 65 and a pallet detection sensor 66. The camera 65 is, for example, a monocular camera. The camera 65 takes images and generates image data. The camera 65 is mounted to take images of the side of the forklift 10.
[0027] The pallet detection sensor 66 is a sensor for detecting pallet P. The pallet detection sensor 66 is a sensor that can detect the position of an object in a three-dimensional coordinate system. The pallet detection sensor 66 can use a sensor similar to the external sensor 61, for example.
[0028] The control device 71 comprises a processor 72 and a storage unit 73. The processor 72 is, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or a DSP (Digital Signal Processor). The storage unit 73 includes RAM (Random access memory) and ROM (Read Only Memory). The storage unit 73 stores program code or instructions configured to cause the processor to execute processing. The storage unit 73, i.e., the computer-readable medium, includes any available medium that can be accessed by a general-purpose or dedicated computer. The control device 71 may be composed of hardware circuits such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). The control device 71, which is a processing circuit, may include one or more processors that operate according to a computer program, one or more hardware circuits such as an ASIC or FPGA, or a combination thereof.
[0029] The control device 71 acquires point cloud data from the external sensor 61. Point cloud data is point data representing the position of an object in a 3D coordinate system. That is, point cloud data is a set of coordinates in the sensor coordinate system. The control device 71 acquires the detection values from the distance meters 62 and 63. The detection values from the distance meters 62 and 63 are the distances to the object detected by the distance meters 62 and 63. The control device 71 acquires the reach amount from the reach amount sensor 64. The control device 71 acquires image data from the camera 65. The control device 71 acquires point cloud data from the pallet detection sensor 66.
[0030] The forklift 10 is equipped with an auxiliary storage device 74. The auxiliary storage device 74 is a non-volatile storage device that can rewrite data. The auxiliary storage device 74 is, for example, a hard disk drive or a solid-state drive.
[0031] The auxiliary storage device 74 stores map data M1 that shows the environment in which the forklift 10 is used. The map data M1 is information about the physical structure of the environment in which the forklift 10 is used, such as the shape of objects in the environment and the size of the environment. In this embodiment, the map data M1 is data that represents the structure of the environment in which the forklift 10 is used in coordinates of a map coordinate system. The map coordinate system is, for example, a two-axis Cartesian coordinate system. The map coordinate system is a coordinate system with an arbitrary point as the origin.
[0032] The control device 71 estimates the self-position of the forklift 10. The self-position is the coordinates of a point on the map coordinate system of the forklift 10. This point on the forklift 10 can be arbitrary, but for example, it is the horizontal center position of the forklift 10.
[0033] Self-position estimation is performed by comparing point cloud data acquired from the external sensor 61 with map data M1. The control device 71 extracts landmarks with the same shape as the landmarks obtained from the point cloud data from the map data M1. The control device 71 recognizes the position of the landmarks from the map data M1. The positional relationship between the position of the landmarks and the forklift 10 can be determined from the detection results of the external sensor 61. Therefore, the control device 71 can estimate its own position by recognizing the position of the landmarks. Landmarks are objects with features that can be identified by the external sensor 61. Landmarks are physical structures whose position does not change easily. Self-position estimation may be performed by combining self-position estimation using the external sensor 61 with dead reckoning using internal sensors. Self-position estimation may be performed by combining self-position estimation using the external sensor 61 with self-position estimation using satellite signals transmitted from GNSS (Global Navigation Satellite System) satellites.
[0034] The control device 71 controls the driving motion. The control device 71 drives the forklift 10 by controlling the drive mechanism 51. The control device 71 controls the cargo handling motion. The control device 71 causes the cargo handling device 21 to perform cargo handling operations by controlling the hydraulic mechanism 52. The cargo handling operations in this embodiment include a lift operation to raise and lower the forks 26, and a reach operation to move the forks 26 forward or backward.
[0035] Forklift 10 performs loading. Loading is the action of placing pallet P, which is loaded onto forks 26, onto truck 100. As shown in Figure 5, the truck 100 is, for example, a wing truck or a flatbed truck. The truck 100 comprises a truck head 101, a cargo bed 102, and two side panels 105 and 106 provided on the sides of the cargo bed 102. The cargo bed 102 is located behind the truck head 101 in the longitudinal direction of the truck 100. The side panels 105 and 106 are provided on both sides of the cargo bed 102. The side panels 105 and 106 include a left side panel 105 provided on the left side of the cargo bed 102 in the lateral direction of the truck 100, and a right side panel 106 provided on the right side of the cargo bed 102 in the lateral direction of the truck 100. The top surface of the cargo bed 102 is a loading surface 103 on which the pallet P is placed. The loading surface 103 is located above the road surface on which the forklift 10 travels.
[0036] As shown in Figure 6, the cargo bed 102 is provided with a stepped portion 104. The stepped portion 104 is located below the loading surface 103. The stepped portion 104 is provided along the side ends E1, which are the ends on both sides of the loading surface 103. When loading is performed, at least one of the side panels 105, 106 is open. That is, at least one of the side panels 105, 106 is rotated below the loading surface 103. When the side panels 105, 106 are open, a gap 107 is created between the cargo bed 102 and the side panels 105, 106.
[0037] Each tailgate 105, 106 has a lower end 108, a side 109, and an upper end 110. The lower end 108 is the end located below when the tailgate 105, 106 is open. The side 109 is the surface facing outwards from the track 100 when the tailgate 105, 106 is open. The upper end 110 is the end located above when the tailgate 105, 106 is open. When the tailgate 105, 106 is open, the lower end 108 is located further outwards from the track 100 than the upper end 110. For the left tailgate 105, the lower end 108 is located to the left of the upper end 110 in the left-right direction of the track 100. For the right tailgate 106, the lower end 108 is located to the right of the upper end 110 in the left-right direction of the track 100.
[0038] When placing pallets P on the loading surface 103, multiple pallets P are arranged side by side. For example, as shown in Figure 5, pallet P1 is placed along the left side 105 and pallet P2 is placed along the right side 106. In this case, the pallet P1 placed along the left side 105 and the pallet P2 placed along the right side 106 are arranged in the front-to-back direction of the truck 100. When placing pallet P1 along the left side 105, the forklift 10 approaches the truck 100 from the left and places pallet P1. When placing pallet P2 along the right side 106, the forklift 10 approaches the truck 100 from the right and places pallet P2.
[0039] Placing multiple pallets P on the loading surface 103 from the same direction constitutes one cycle. "Same direction" means that when placing pallets P on the loading surface 103, the forklift 10 approaches the loading surface 103 from the same direction. In the example shown in Figure 5, one cycle involves approaching the truck 100 from the left and placing three pallets P1. Another cycle involves approaching the truck 100 from the right and placing three pallets P2.
[0040] This section describes the cargo handling control performed by the control device 71 when the forklift 10 places cargo. The cargo handling control includes control performed when the pallet P is placed for the first time, and control performed when the pallet P is placed for the second time or later. Once one cycle is completed, the cargo handling control ends. That is, the following cargo handling controls are performed within the same cycle.
[0041] <When placing the palette for the first time> As shown in Figure 7, in step S1, the control device 71 detects the truck 100. The forklift 10 travels along the side panels 105 and 106 of the truck 100 in the longitudinal direction of the truck 100. The control device 71 detects the truck 100 using point cloud data acquired from the external sensor 61 while the forklift 10 is traveling. For example, the control device 71 detects the mounting surface 103 from the point cloud data. The mounting surface 103 can be detected, for example, by a method of extracting a horizontal plane from the point cloud data. The method for detecting the truck 100 is arbitrary. For example, the control device 71 may detect the truck 100 from image data acquired from the camera 65 by using object recognition using machine learning.
[0042] Next, in step S2, the control device 71 generates a path. The path indicates the position and orientation of the forklift 10. As shown in Figure 5, the generated path R is the path to the loading position set on the loading surface 103. Path R includes a first path R1 extending along the side panels 105, 106 of the truck 100, and a second path R2 extending from the first path R1 toward the loading position. The first path R1 includes a path Rs extending in the front-rear direction of the truck 100, and a path Rc curving from path Rs toward the second path R2. The second path R2 is the path extending in the left-right direction of the truck 100. The loading position is set at a location on the loading surface 103 where a pallet P can be placed. When placing pallets P sequentially from the front to the rear of the truck 100, the loading position may be set as close as possible to the truck head 101. The loading position is, for example, a position that coincides with the center of the front of the pallet P when the pallet P is placed on the loading surface 103. That is, the loading position is the target position of the center of the front of the pallet P. The front of pallet P is the surface that faces the forklift 10 when pallet P is being transported by the forklift 10.
[0043] Next, in step S3, the control device 71 performs an approach. The approach is the process of moving the forklift 10 to the front of the loading position. The control device 71 controls the drive mechanism 51 while knowing its own position, causing the forklift 10 to travel so that it follows the first path R1.
[0044] Next, in step S4, the control device 71 moves the forklift 10 forward. The control device 71 controls the drive mechanism 51 while knowing its own position, causing the forklift 10 to travel along the second path R2. As a result, the forklift 10 approaches the loading position.
[0045] Next, in step S5, the control device 71 performs angular misalignment detection. Angular misalignment detection means calculating the misalignment angle θ1 between the mounting surface 103 and the pallet P. The misalignment angle θ1 is the angle of the front surface of the pallet P with respect to the longitudinal direction of the track 100. The misalignment angle θ1 is caused by a misalignment in the entry angle of the forks 26 relative to the mounting surface 103. When the forks 26 are moved above the mounting surface 103, a misalignment angle θ1 occurs when the two forks 26 enter at an angle relative to the side end E1.
[0046] As shown in Figure 9, the deviation θ2 of the entry angle of the fork 26 relative to the mounting surface 103 is the angle of the line segment L1 relative to the side end E1. The line segment L1 is the line segment connecting the tips of the two support parts 27. When the fork 26 enters perpendicular to the side end E1, the deviation θ2 of the entry angle of the fork 26 is 0. The deviation θ2 of the entry angle of the fork 26 relative to the mounting surface 103 can be considered the same as the deviation angle θ1. Therefore, by detecting the deviation θ2 of the entry angle of the fork 26 relative to the mounting surface 103, the deviation angle θ1 can be detected. Angle deviation detection is performed using the detected values of the first distance meter 62 and the second distance meter 63. Angle deviation detection will be explained in detail below.
[0047] As shown in Figure 8, in step S11, the control device 71 raises the support portion 27 to a position higher than the mounting surface 103 and moves the forks 26 forward toward the loading position. For example, the control device 71 moves the mast 22 forward by the reach cylinder 32 by controlling the hydraulic mechanism 52. The control device 71 may also move the forklift 10 forward by controlling the drive mechanism 51.
[0048] Next, in step S12, the control device 71 acquires the detected values of the distance meters 62 and 63 while the fork 26 is moving forward. As the fork 26 passes over the side panels 105 and 106 and the cargo bed 102, the first distance meter 62 and the second distance meter 63 detect distances that conform to the shapes of the side panels 105 and 106 and the cargo bed 102.
[0049] As shown in Figure 10, before the distance meters 62 and 63 reach above the tailgates 105 and 106, the distance meters 62 and 63 detect the distance to the ground. At the location where the truck 100 stops, the ground is often relatively flat, and the change in distance detected by the distance meters 62 and 63 is minimal while the distance to the ground is being detected.
[0050] At the first time point t1, when the distance meters 62 and 63 reach above the lower end 108 of the tailgates 105 and 106, the distance meters 62 and 63 detect the distance to the tailgates 105 and 106. The distance detected by the distance meters 62 and 63 is shorter than when detecting the distance to the ground. As the fork 26 moves forward, the distance meters 62 and 63 detect the distance following the slope of the side 109 of the tailgates 105 and 106. The tailgates 105 and 106 are sloped so that they move away from the cargo bed 102 from the upper end 110 to the lower end 108. Therefore, as the fork 26 moves forward, the distance meters 62 and 63 move closer to the upper end 110 of the tailgates 105 and 106 from the lower end 108, and the distance detected by the distance meters 62 and 63 decreases.
[0051] At the second time point t2, when the distance meters 62 and 63 reach above the upper end 110, they detect the distance to the upper end 110. With the tailgates 105 and 106 open, the upper end 110 has a gentler slope compared to the side 109. Therefore, the change in distance detected by the distance meters 62 and 63 becomes smaller after the second time point t2.
[0052] At the third time point t3, when distance meters 62 and 63 reach above gap 107, they detect the distance to the ground. As a result, the distance detected by distance meters 62 and 63 becomes longer.
[0053] At the fourth time point t4, when the distance meters 62 and 63 reach above the stepped section 104, they detect the distance to the stepped section 104. Since the stepped section 104 is located above ground level, the distance detected by the distance meters 62 and 63 is shorter. Because the stepped section 104 has little elevation change, the change in the distance detected by the distance meters 62 and 63 is minimal while detecting the distance to the stepped section 104. Thus, the detected value by the distance meters 62 and 63 changes depending on the positional relationship between the distance meters 62 and 63 and the track 100.
[0054] Next, in step S13, the control device 71 detects a first position time and a second position time. Based on the change in the detected value of the first distance meter 62 as the fork 26 moves forward, the control device 71 detects a first position time when the first distance meter 62 is positioned above the upper ends 110 of the tailgates 105 and 106. Based on the change in the detected value of the second distance meter 63 as the fork 26 moves forward, the control device 71 detects a second position time when the second distance meter 63 is positioned above the upper ends 110 of the tailgates 105 and 106. For example, the amount of change in distance detected by the distance meters 62 and 63 decreases after the second time t2. Therefore, the first position time can be detected by setting a threshold for the difference between the current value and the previous value of the detected value of the first distance meter 62, or by calculating a slope that represents the change in the detected value. Similarly, the second position time can be detected by setting a threshold for the difference between the current value and the previous value of the detected value of the second distance meter 63, or by calculating a slope that represents the change in the detected value.
[0055] Next, in step S14, the control device 71 calculates the displacement angle θ1 between the mounting surface 103 and the pallet P1, i.e., the displacement θ2 of the entry angle of the fork 26, based on the distance X1 traveled by the fork 26 between the first position time and the second position time. The smaller the displacement θ2 of the entry angle of the fork 26, the shorter the travel distance X1. The control device 71 calculates the elapsed time from the earlier of the first position time and the later of the second position time. The control device 71 calculates the travel distance X1 as the product of the calculated elapsed time and the travel speed of the fork 26. Then, the control device 71 calculates the displacement θ2 of the entry angle from equation (1).
[0056] The deviation in the entry angle θ² = sin -1 (Movement distance X1 / Separation distance Z)…(1) The separation distance Z is the distance between the first distance meter 62 and the second distance meter 63. Since the separation distance Z is a known value, it can be stored in advance in a memory medium that the control device 71 can read. The travel speed of the forks 26 is the travel speed of the forklift 10 when the forks 26 are moved by the movement of the forklift 10. The travel speed of the forks 26 is the travel speed of the mast 22 when the forks 26 are moved by a reach operation. The travel speed of the forklift 10 can be obtained from the speedometer. The travel speed of the mast 22 can be calculated from the detected value of the reach sensor 64.
[0057] The control device 71 may calculate the travel distance X1 by odometry. For example, the control device 71 starts odometry measurement when it detects either the first position time or the second position time. Once odometry measurement is started, the control device 71 accumulates the amount of movement of the fork 26. For example, the control device 71 may accumulate the reach amount obtained from the reach amount sensor 64. The control device 71 may also accumulate the amount of movement using an encoder that detects the rotational speed of the front wheel 13. The control device 71 ends odometry measurement when it detects the other of the first position time or the second position time. As a result, the amount of movement accumulated from when one of the two distance meters 62, 63 is above the upper end 110 until the other is above the upper end 110 is calculated as the travel distance X1.
[0058] As shown in Figure 7, in step S6, the control device 71 detects the edge of the cargo bed. The edge of the cargo bed is the upper end 110 of the side panels 105, 106, or the stepped portion 104. The edge of the cargo bed can be detected using either the detection value of the first distance meter 62 or the detection value of the second distance meter 63. The control device 71 may use the detection value of the first distance meter 62 to detect the edge of the cargo bed, or it may use the detection value of the second distance meter 63 to detect the edge of the cargo bed. As an example, the case in which the upper end 110 of the side panels 105, 106 is detected using a first linear function f1 and a second linear function f2 will be described.
[0059] The first linear function f1 is a linear function that represents the change over time of the detected values obtained from the time the distance meters 62 and 63 began detecting the distance to tilt 105 and 106 until the second timing, which is a predetermined time before the first timing when the latest detected value was obtained. The second linear function f2 is a linear function that represents the change over time of the detected values obtained until the first timing, from the second timing onward. The first linear function f1 and the second linear function f2 can be calculated, for example, by the least squares method.
[0060] While the distances to the sides 109 of the tailgates 105 and 106 are detected by the distance meters 62 and 63, the slopes of the first linear function f1 and the second linear function f2 are identical. When the distances to the upper ends 110 of the tailgates 105 and 106 are detected by the distance meters 62 and 63, the change in the detected values of the distance meters 62 and 63 becomes small, causing a difference to occur between the slopes of the first linear function f1 and the second linear function f2.
[0061] The control device 71 identifies the intersection point of the first linear function f1 and the second linear function f2 when the difference between the slopes of the first linear function f1 and the second linear function f2 exceeds a predetermined value. The control device 71 identifies that the positions of the distance meters 62 and 63 at the timing of the identified intersection point are the positions of the upper ends 110 of the tilt beams 105 and 106. The predetermined value is set so that the difference between the slopes of the first linear function f1 and the second linear function f2 can be determined by detecting the upper ends 110 of the tilt beams 105 and 106.
[0062] Next, in step S7, the control device 71 stops the forks 26. If the forklift 10 is moving, the control device 71 stops the forklift 10. If the mast 22 is moving forward by the reach operation, the control device 71 stops the reach operation.
[0063] Next, in step S8, the control device 71 performs a turn in place of the forklift 10 so that the deviation θ2 of the entry angle falls within the allowable range. For example, the control device 71 performs a turn in place by the amount of the deviation θ2 of the entry angle detected in step S5. If the forks 26 are tilted to the left relative to the loading platform 102, the control device 71 turns the forklift 10 to the right by the amount of the deviation θ2 of the entry angle. If the forks 26 are tilted to the right relative to the loading platform 102, the control device 71 turns the forklift 10 to the left by the amount of the deviation θ2 of the entry angle. Turning in place is an action in which the two front wheels 13 and rear wheels 14 are used as steering wheels and the forklift turns in place, that is, turns in a circular motion. The control device 71 does not have to perform a turn in place if the deviation θ2 of the entry angle detected in step S5 falls within the allowable range. The allowable range can be set arbitrarily.
[0064] Next, in step S9, the control device 71 performs a cargo handling operation. The control device 71 places the pallet P at the loading position by operating the cargo handling device 21. For example, the control device 71 places the pallet P at a predetermined distance from the end of the loading platform in the depth direction of the loading platform 102. The predetermined distance is set so that the pallet P does not interfere with the side panels 105 and 106 when the pallet P is placed at the loading position. The control device 71 places the pallet P at the loading position by advancing the forks 26 so that the pallet P is positioned above the loading position, and then lowering the forks 26.
[0065] <When placing a palette for the second time or later> The control performed by the control device 71 when placing pallet P for the second time or later will be explained. As shown in Figure 11, in step S21, the control device 71 detects the pallet P. Detecting the pallet P involves detecting the pallet P already placed on the mounting surface 103. The control device 71 detects the pallet P using the detection results from the camera 65 and the detection results from the pallet detection sensor 66. The control device 71 acquires image data from the camera 65. Based on the image data, the control device 71 recognizes the pallet P. The control device 71 recognizes the pallet P using, for example, image processing technology utilizing machine learning.
[0066] The control device 71 extracts point cloud data corresponding to pallet P from the point cloud data acquired from the pallet detection sensor 66. This enables detection of pallet P. The position of pallet P is, for example, the 3D coordinate of the center of the front surface of pallet P. The position of the external sensor 61 and the position of the pallet detection sensor 66 are different. The control device 71 may convert the position of pallet P with the pallet detection sensor 66 as the origin to the position of pallet P with the external sensor 61 as the origin, based on the difference between the position of the external sensor 61 and the position of the pallet detection sensor 66.
[0067] Next, in step S22, the control device 71 generates a path R. When placing a pallet P for the second time or later, the placement position may be a predetermined distance away from the pallets P already placed on the loading platform 102 in the longitudinal direction of the truck 100. For example, the placement position is a predetermined distance away from the rear of the truck 100 from the center of the front of the rearmost pallet P already placed on the loading platform 102. The predetermined distance is set so that adjacent pallets P do not interfere with each other when placed at the placement position. The predetermined distance is, for example, the width of the pallet P plus a margin.
[0068] The path R generated in step S22 is the path R corrected by the deviation θ2 of the approach angle detected when the pallet P was placed for the first time, i.e., the deviation angle θ1. The control device 71 generates the path R by adding or subtracting the deviation θ2 of the approach angle detected in step S5 from the attitude.
[0069] Next, in step S23, the control device 71 makes an approach. Step S23 is the same process as in step S3. Next, in step S24, the control device 71 moves the forklift 10 forward. Step S24 is the same process as in step S4.
[0070] Next, in step S25, the control device 71 detects the angular displacement of the pallet P. Step S25 is the same process as in step S5. Next, in step S26, the control device 71 detects the edge of the loading platform. Step S26 is the same process as in step S5.
[0071] Next, in step S27, the fork 26 is stopped. Step S27 is the same process as in step S7. Next, in step S28, the control device 71 performs an in-situ turn of the forklift 10 so that the deviation θ2 of the entry angle falls within the allowable range. Step S28 is the same process as in step S8. In step S28, the forklift 10 performs an in-situ turn by the amount of the deviation θ2 of the entry angle detected in step S25.
[0072] Next, in step S29, the control device 71 performs a cargo handling operation. Step S29 is the same process as in step S9. If the displacement angle θ1 detected in step S25 is greater than the displacement angle θ1 detected in step S5, the control device 71 may cancel the correction of path R due to the displacement angle θ1 when generating path R in subsequent cycles. If the displacement angle θ1 detected in step S25 is greater than the displacement angle θ1 detected in step S5 for multiple consecutive cycles, the control device 71 may cancel the correction of path R due to the displacement angle θ1 when generating path R in subsequent cycles. If the correction of path R due to the displacement angle θ1 is canceled, the correction of path R does not need to be performed in the cycle in which the cancellation occurred. If the correction of path R is canceled, the control device may treat the pallet P to be placed next as the pallet P to be placed first.
[0073] [Effect of the Embodiment] The control device 71 calculates the deviation θ2 of the approach angle based on the detected values of the two distance meters 62 and 63. When the two distance meters 62 and 63 are configured to detect the distance to an object below, the detected values of the two distance meters 62 and 63 change in accordance with the shape of the track 100. Therefore, it is possible to understand the characteristic parts of the track 100 from the detected values of the two distance meters 62 and 63, and to calculate the deviation θ2 of the approach angle from the time difference in which the two distance meters 62 and 63 are positioned above the characteristic parts.
[0074] In this embodiment, the deviation θ2 of the entry angle is calculated from the difference between a first position where the first distance meter 62 is located above the upper ends 110 of the tailgates 105 and 106, and a second position where the second distance meter 63 is located above the upper ends 110 of the tailgates 105 and 106.
[0075] At the first time step t1, when the distance meters 62 and 63 are positioned above the lower end 108 of the side panels 105 and 106, the distance detected by the distance meters 62 and 63 is shorter than when the distance to the ground is detected. For this reason, it is also possible to calculate the deviation θ2 of the entry angle using the lower end 108 of the side panels 105 and 106 as a feature part. However, when the lower end 108 of the side panels 105 and 106 is used as a feature part, the distance to the loading position is long, which may lead to a large discrepancy between the deviation θ2 of the entry angle when the load is actually placed and the calculated deviation θ2 of the entry angle.
[0076] At the third time step t3, when the distance meters 62 and 63 are positioned above the gap 107, the distance detected by the distance meters 62 and 63 is longer than when they are detecting the distance to the side 109. Therefore, it is possible to calculate the deviation θ2 of the entry angle using the gap 107 as a feature. However, if there is an obstacle blocking the gap 107, the distance meters 62 and 63 may not be able to capture the features of the gap 107, and the deviation θ2 of the entry angle may not be able to be calculated.
[0077] At the fourth time step t4, when the distance meters 62 and 63 are positioned above the stepped section 104, the distance detected by the distance meters 62 and 63 is shorter than when the distance to the ground is detected. Therefore, it is possible to calculate the deviation θ2 of the entry angle using the stepped section 104 as a feature. However, if there are obstacles such as hinges that rotate the tailgates 105 and 106 near the stepped section 104, the distance meters 62 and 63 may not be able to capture the features of the stepped section 104, and the deviation θ2 of the entry angle may not be calculated.
[0078] In contrast, if an obstacle is present at the upper end 110 of the tailgate 105, 106, the tailgate 105, 106 cannot be closed, so obstacles are rarely placed at the upper end 110 of the tailgate 105, 106. Also, since the upper end 110 of the tailgate 105, 106 is closer to the cargo placement position than the lower end 108 of the tailgate 105, 106, the discrepancy between the deviation θ2 of the entry angle when actually placing the cargo and the calculated deviation θ2 of the entry angle is less likely to be large.
[0079] The control device 71 then corrects the path R1 based on the calculated deviation θ2 of the entry angle. As shown in Figure 5, when the forklift 10 places multiple pallets P from the same direction, the deviation angle θ1 between the placement surface 103 and each pallet P tends to be similar. For example, the deviation angle θ1 of each pallet P1 tends to be similar. The deviation angle θ1 of each pallet P2 tends to be similar.
[0080] The cause of the deviation angle θ1 could be, for example, individual differences in the forklift 10, the accuracy of sensors such as the external sensor 61 and camera 65, or a decrease in the accuracy of self-position estimation due to the surrounding environment of the truck 100. Furthermore, even if the path R itself is accurate, the deviation angle θ1 will increase if the forklift 10 cannot accurately follow the path R, or if the path R itself is not accurately generated.
[0081] The above-mentioned causes are not accidental but always present; therefore, when the forklift 10 places multiple pallets P from the same direction, the displacement angle θ1 between the placement surface 103 and each pallet P tends to be similar. In other words, multiple pallets P tend to tilt in the same way. To put it another way, within the same cycle, the displacement angle θ1 can be considered identical.
[0082] [Effects of the Embodiment] (1) The control device 71 detects a first position time when the first distance meter 62 is above the upper ends 110 of the tailgates 105 and 106, based on the change in the detected value of the first distance meter 62 as the forks 26 move forward. The control device 71 detects a second position time when the second distance meter 63 is above the upper ends 110 of the tailgates 105 and 106, based on the change in the detected value of the second distance meter 63 as the forks 26 move forward. The control device 71 can then calculate the displacement angle θ1 based on the distance X1 traveled by the forks 26 between the first position time and the second position time. By calculating the displacement angle θ1 using the upper ends 110 of the tailgates 105 and 106 as a feature portion, it is less likely that the displacement angle θ1 cannot be calculated due to the influence of obstacles. In addition, the discrepancy between the displacement angle θ1 when the load is actually placed and the calculated displacement angle θ1 is less likely to be large.
[0083] (2) The control device 71 corrects the path R based on the displacement angle θ1 calculated in the past within the same cycle. When the forklift 10 places multiple pallets P from the same direction, the displacement angle θ1 between the placement surface 103 and each pallet P tends to be a similar angle. Therefore, by correcting the path R based on the displacement angle θ1 calculated in the past, the displacement angle θ1 can be reduced when placing pallets P from the second time onward.
[0084] (3) The control device 71 rotates the forklift 10 in place by the amount of the displacement angle θ1, and then places the pallet P. When placing the pallet P for the first time, the path R cannot be corrected, so rotating in place makes the displacement angle θ1 smaller.
[0085] When placing pallet P for the second time or later, although the path R is corrected, the deviation angle θ1 may exceed the acceptable range. Therefore, even for the second time and beyond, the deviation angle θ1 can be reduced by performing a rotation in place.
[0086] (4) When the control device 71 places the pallet P on the mounting surface 103 for the second time or later, it corrects the path R based on the displacement angle θ1 calculated when the pallet P was placed on the mounting surface 103 for the first time. When placing the pallet P for the second time or later, the forklift 10 travels along the corrected path R, so it is not possible to detect the displacement angle θ1 when the path R is not corrected. That is, the displacement angle θ1 may be smaller than when the path R is not corrected. By correcting the path R based on the displacement angle θ1 calculated when the pallet P was placed on the mounting surface 103 for the first time, the displacement angle θ1 when it is not corrected can be used. Therefore, the displacement angle θ1 can be appropriately reduced.
[0087] (5) The displacement angle θ1 can be calculated using distance meters 62 and 63 for detecting the edge of the loading platform. Since it is not necessary to use a dedicated sensor to calculate the displacement angle θ1, the increase in the number of parts can be suppressed.
[0088] [Example of changes] The embodiment can be implemented with the following modifications. The embodiment and the following modifications can be combined with each other to the extent that they do not contradict each other technically.
[0089] ○The control device 71 may stop loading if the calculated displacement angle θ1 exceeds the allowable range. ○The control device 71 may retry if the calculated deviation angle θ1 exceeds the allowable range. A retry means executing the loading operation again. For example, if the calculated deviation angle θ1 exceeds the allowable range, the control device 71 may reverse the forklift 10 once and then execute the loading control operation again. In this case, the control device 71 may correct the path R based on the calculated deviation angle θ1, or it may rotate in place by the amount of the calculated deviation angle θ1 before executing the loading control operation. The control device 71 may retry until the deviation angle θ1 falls within the allowable range. In this way, the control device 71 can perform various controls according to the calculated deviation angle θ1.
[0090] ○The control device 71 may correct the path R when placing the pallet P for the third time or later, based on the displacement angle θ1 calculated when placing the pallet P on the mounting surface 103 for the second time or later within the same cycle.
[0091] ○The detection of the first and second position times may be performed using the difference in the moving average of the detected values from distance meters 62 and 63. As shown in Figure 12, when calculating the difference in the moving average of the detected values from distance meters 62 and 63, the difference in the moving average is 0 up to the first time step t1. The number of detected points used when calculating the moving average is arbitrary. The difference value is the difference between the previous value of the moving average and the current value of the moving average. For example, the difference value is the current value of the moving average minus the previous value of the moving average. When distance meters 62 and 63 are detecting the distance to the ground, the change in distance is small, so the difference in the moving average will be 0 or close to 0.
[0092] At the first time point t1, when distance meters 62 and 63 begin to detect the distance to the side 109 of the tilt surfaces 105 and 106, a change occurs in the distance detected by distance meters 62 and 63, causing the difference value of the moving average to become less than 0. As long as the moving average includes the distance to the ground, the difference value of the moving average fluctuates significantly due to the difference between the distance to the ground and the distance to the tilt surfaces 105 and 106. Once the moving average no longer includes the distance to the ground, the change in the difference value of the moving average becomes more gradual.
[0093] As the second time point t2 arrives and the distance meters 62 and 63 begin to detect the distance to the upper end 110 of the tilt 105 and 106, the change in the detected values of the distance meters 62 and 63 decreases. As a result, the difference value approaches 0.
[0094] At the third time point, t3, when the distance to the ground is detected, the distance detected by distance meters 62 and 63 increases. As a result, the distance to the ground is included in the moving average, and the difference value of the moving average becomes greater than 0.
[0095] When the fourth time step, t4, is reached and the distance to the stepped section 104 is detected, the distance detected by the distance meters 62 and 63 becomes shorter. As a result, the distance to the stepped section 104 is included in the moving average, and the difference value of the moving average becomes less than 0.
[0096] Thus, even when using the difference value of a moving average, the difference value fluctuates significantly in the feature portion. For this reason, the control device 71 can detect the first position time and the second position time from the difference value.
[0097] ○The external sensor 61 may also be used as the pallet detection sensor 66. ○Detection of the edge of the loading platform may be performed using point cloud data acquired from the external sensor 61 or the pallet detection sensor 66.
[0098] ○Detection of pallet P may be performed using the external sensor 61. For example, the control device 71 extracts point cloud data corresponding to pallet P from the point cloud data acquired from the external sensor 61. The point cloud data corresponding to pallet P is obtained, for example, by matching the shape of pallet P with the point cloud data acquired from the external sensor 61.
[0099] ○The control device 71 may, instead of turning in place, turn while moving to cancel out the displacement angle θ1. In this case, if the distance between the forklift 10 and the truck 100 is short and the necessary distance to turn cannot be secured, the control device 71 may secure the distance by having the forklift 10 move in reverse.
[0100] The distance meters 62 and 63 may be two-dimensional distance meters 62 and 63 that detect distances in two directions. For example, the distance meters 62 and 63 may detect distances in two directions by irradiating a laser in a fan shape.
[0101] ○Forklift 10 may be a counterbalanced forklift. ○Self-position estimation, driving motion, and cargo handling motion may each be controlled by different control devices. [Explanation of symbols]
[0102] P...Pallet, an example of a load; R...Route; 10...Forklift; 26...Fork; 62...First distance meter; 63...Second distance meter; 71...Control device; 100...Truck; 102...Loading bed; 103...Loading surface; 105, 106...Side panels; 110...Top edge.
Claims
1. A forklift for loading cargo onto a truck equipped with side panels on the sides of the cargo bed, Two forks are positioned spaced apart from each other in the left-right direction, A first distance meter that moves in conjunction with the two forks and detects the distance to an object below the two forks, A second distance meter that moves in conjunction with the two forks and detects the distance to an object below the two forks, the second distance meter being positioned separately from the first distance meter in the left-right direction, A control device is provided, The control device is The two forks are advanced so that they pass above the tailgate. Based on the change in the detected value of the first distance meter as the two forks move forward, a first position is detected in which the first distance meter is located above the upper end of the tailgate. Based on the change in the detected value of the second distance meter as the two forks move forward, a second position is detected in which the second distance meter is located above the upper end of the tailgate. A forklift that calculates the displacement angle between the loading surface, which is the upper surface of the loading platform, and the load, based on the distance traveled by the two forks between the first position and the second position.
2. The control device is A path is generated to the loading position set on the aforementioned loading surface. Move the forklift along the aforementioned route and place the load on the aforementioned surface. If we consider placing multiple loads on the aforementioned mounting surface from the same direction as one cycle, then, The control device is When placing the load on the aforementioned surface for the first time in the aforementioned cycle, the displacement angle is calculated. The forklift according to claim 1, wherein when placing the load on the aforementioned surface for the second time or later in the aforementioned cycle, the path is corrected based on the displacement angle previously calculated within the same cycle.
3. The control device rotates the forklift in place by the amount of the displacement angle, and then places the load on it, as described in claim 1.
Citation Information
Patent Citations
Position detector, cargo handling vehicle, and program
JP2023142353A