Multi-station camera positioning system

Abstract

The invention relates to the technical field of industrial automation, and discloses a multi-station visual positioning feeding system. The system comprises a movable visual positioning unit, a multi-station material frame bearing area, a robot execution unit and a control system. According to the main technical scheme, a horizontal moving mechanism drives a visual unit to scan at least two material frame stations, and workpiece real-time pose data are generated; the control system controls the robot to continuously grab other material frame workpieces when a certain material frame is replaced on the basis of state collaborative logic; wherein the material frame in-place sensor triggers special scanning of a new material frame, and the state machine achieves automatic switching of a full material state, a to-be-replaced state in operation. The method has the beneficial effects that 1, zero shutdown is realized in the material changing process, and the equipment utilization rate is increased to 95% or above; 2, 3D vision closed-loop positioning enables the grabbing precision of the scattered workpieces to reach + / -0.1 mm, and the success rate is greater than 99.5%; the automatic feeding device is suitable for automatic feeding of bulk materials such as bearing rollers and gear blanks.

Description

Technical Field

[0001] This invention relates to the field of industrial automation technology, and in particular to a multi-station camera positioning system. Background Technology

[0002] In the automated production of precision parts such as bearings and gears, there is a significant technical bottleneck in the workpiece loading process:

[0003] 1. Inefficient manual material loading mode: incoming materials are usually stored in the material box in a loose pile, which requires operators to manually pick up and place them on the conveyor line. This is labor-intensive and slow (about 15-20 pieces / minute). Manual operation is prone to surface contamination or damage to the workpiece, which affects the product yield (industry average loss rate ≥3%).

[0004] 2. Low utilization rate of single-frame automatic feeding system: Most existing automated equipment adopts a single-station frame design. When the frame is full of workpieces, the machine must be stopped and wait for manual replacement. The replacement process takes an average of 30-60 seconds, resulting in an equipment idle rate of over 40% (based on actual measurement data from an automotive parts factory).

[0005] 3. Insufficient positioning accuracy of disordered workpieces: The random pose of scattered workpieces causes traditional mechanical positioning to fail, and the failure rate of some systems using fixed trajectory robots is >8%; two-dimensional vision positioning cannot adapt to highly overlapping and stacked scenarios, especially the recognition error of reflective metal parts (such as bearing rollers) often exceeds ±1mm.

[0006] 4. Collaboration deficiencies of multi-station systems: Although there is a parallel design of multiple material frames, there is a lack of status awareness and dynamic scheduling mechanism, and the overall operation still needs to be paused when changing materials; the fixed installation of vision units leads to scanning blind spots, making it difficult to cover the entire workpiece area of ​​material frames of different sizes.

[0007] In summary, existing technologies cannot simultaneously address the three core requirements of efficient continuous feeding (eliminating downtime during material changes), high-precision disordered grasping (adapting to scattered workpieces), and intelligent multi-station collaboration. Summary of the Invention

[0009] To overcome the above-mentioned deficiencies of the prior art, embodiments of the present invention provide a multi-position camera positioning system.

[0010] To achieve the above objectives, the innovative aspects of this invention are as follows: its structure includes a mobile vision positioning unit, a multi-station material frame carrying area, a robot execution unit, and a control system; the multi-station material frame carrying area has at least two independent material frame stations; the mobile vision positioning unit moves above each material frame station through a horizontal moving mechanism and performs 3D vision positioning on the scattered workpieces within the material frame station, generating real-time updated workpiece positioning data; the control system generates grasping instructions based on the workpiece positioning data, controlling the robot execution unit to grasp workpieces in different material frame stations across stations; wherein, when a certain material frame station is in an empty frame replacement state, the robot execution unit continues to grasp workpieces based on the pre-stored positioning data of the other material frame stations.

[0011] Furthermore, the aforementioned horizontal moving mechanism includes a support column, a linear guide rail, and a drive motor; the linear guide rail is horizontally fixed on the support column and spans the multi-station material frame bearing area; a sliding bracket is provided on the linear guide rail; a mobile vision positioning unit is vertically mounted on the bracket; the drive motor drives the bracket to move horizontally along the linear guide rail through a transmission component.

[0012] Furthermore, the aforementioned drive motor is a servo motor and is equipped with a position detection element, and the transmission component is a precision ball screw; the servo motor, precision ball screw, and position detection element constitute a closed-loop control system with a horizontal positioning accuracy ≤0.1mm.

[0013] Furthermore, the aforementioned mobile visual positioning unit is a 3D camera.

[0014] Furthermore, the aforementioned multi-station material frame carrying area is equipped with three material frame stations, arranged linearly or in a circular array; each material frame station is equipped with a material frame positioning sensor; the material frame positioning sensor is used to detect whether the material frame is in an operable position.

[0015] Furthermore, the aforementioned robot execution unit is a six-axis industrial robot with a vacuum suction cup or adaptive gripper installed at the end.

[0016] Furthermore, the aforementioned mobile vision positioning unit executes a dynamic scanning strategy: a) Initial scan: sequentially scans all material frame stations to generate global workpiece coordinates; b) Incremental scan: when the change in the number of workpieces at a certain material frame station exceeds a set value, a local rescan is triggered; c) New frame scan: when the material frame positioning sensor detects that a new material frame is in place and its status is "to be replaced", a special scan of that material frame station is triggered.

[0017] Furthermore, the multi-station collaborative logic of the above control system includes: (1) establishing a status identifier for each material frame station, including full status, working status and waiting-to-be-replaced status; (2) when the status of a material frame station is switched to waiting-to-be-replaced status: freezing the grabbing instruction of the material frame station and activating the pre-stored positioning data of the other material frame stations; (3) after receiving the replacement completion signal from the material frame positioning sensor, updating the status identifier and triggering the new frame scanning.

[0018] Furthermore, the switching rules for the above status indicators include: switching from full status to working status: triggered when the material frame station is grabbed for the first time; switching from working status to waiting-to-replace status: triggered when the visual scan detects that the number of workpieces is lower than the set threshold or the number of consecutive grabbing failures reaches the fault tolerance limit; switching from waiting-to-replace status to full status: triggered after receiving the material frame positioning sensor signal.

[0019] Furthermore, the above system is applied to automated feeding scenarios for bearing rollers, gear blanks, and injection molded parts.

[0020] The beneficial effects of this invention are:

[0021] 1. Achieve zero downtime during material changeover and increase equipment utilization to over 95%: A multi-station status collaboration mechanism monitors the material frame status in real time; the arrival sensor triggers a special scan of the new material frame; during material changeover, the robot automatically switches to other stations for continuous operation, reducing the traditional system's 40% idle rate to <5%.

[0022] 2. The accuracy of grasping scattered workpieces reaches ±0.1mm, with a success rate of >99.5%: The mobile 3D camera generates real-time point clouds through full-domain scanning; the positioning accuracy of the servo closed-loop control system is ≤0.1mm; the success rate of grasping typical workpieces such as bearing rollers and gear blanks is ≥99.5%, which is 14.2 percentage points higher than that of traditional systems. Attached Figure Description

[0024] Figure 1 This is an isometric drawing of the present invention.

[0025] Figure 2 Flowchart of multi-station status switching. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] , like Figure 1 As shown, the system hardware implementation

[0030] 1. Construction of multi-station material frame support area: A steel structure platform with a length of 2000 mm, a width of 1500 mm, and a height of 800 mm is adopted, and the platform surface is covered with anti-static rubber mats; three independent material frame stations are set up, with a center-to-center distance of 600 mm between the stations; photoelectric beam sensors are installed at the four corners of the base of each station, and the optical axis of the sensor is 10 mm above the platform surface. When the material frame completely blocks the four beams, the sensor outputs a high-level ready signal to the control system.

[0031] 2. Installation of the horizontal moving mechanism 2. A support column 21 with a cross-section of 200 mm × 200 mm is fixed to the side of the bearing area, and the support column 21 has a height of 2500 mm; a linear guide rail 22 is horizontally fixed to the top of the support column 21, and the linear guide rail 22 has a length of 1800 mm to cover the three work positions; an aluminum alloy bracket 3 is connected to the linear guide rail 22 via a slider; a movable vision positioning unit 1 is installed on the bracket 3 via a lifting assembly, with a lifting stroke of 0-300 mm, and the height can be adjusted by manually tightening bolts or a servo electric cylinder.

[0032] 3. Implementation of Closed-Loop Control System: The drive motor 23 is a 200-watt servo motor, which drives the support 3 to move via a precision ball screw with a lead of 10 mm. The position detection elements include a 17-bit absolute encoder at the motor end and a grating ruler at the guide rail end, with a resolution of 0.1 micrometers. The control system reads the target position and the encoder feedback position in real time, and dynamically adjusts the motor speed using a proportional-integral-derivative algorithm until the position error is less than 0.05 mm. Tested with a laser tracker, the maximum positioning error after 50 rounds of motion is 0.08 mm.

[0033] 4. Robot Actuator Configuration: Six-axis industrial robot with repeatability of ±0.02 mm and load capacity of 35 kg; the end effector is selected from the following: Vacuum suction cup assembly: suitable for smooth bearing rollers; Adaptive gripper: built-in pressure feedback, suitable for irregular gear blanks.

[0034] 5. Visual positioning unit implementation: 3D camera resolution 1280×1024 pixels; positioning process: camera projects coded stripes → binocular image acquisition → stereo matching to generate disparity map → 3D coordinate transformation → output workpiece pose data; verified using a standard ball with a diameter of 20.00 mm, measurement error ±0.03 mm.

[0035] 6. The workpiece quantity threshold achieves ±2% accuracy through point cloud volume calculation, specifically including: a) Calculating the current total point cloud volume: Generating the point cloud outer contour using the convex hull algorithm and calculating the containment volume using the voxel mesh method; b) Estimating the current number of workpieces: ,in, : Current total volume of point cloud (mm³); : Workpiece stacking density coefficient (0.95 for bearing rollers); c) Standard volume of a single piece (obtained through calculation using a CAD model); d) Dynamic calibration: Every 5 scans, a standard sample (known quantity) is taken for error compensation and updated. value.

[0036] Control logic implementation

[0037] 1. Dynamic scanning strategy: Initial scan: The camera moves in the order of station A→B→C, and multiple sets of point cloud data are collected and fused at each station; Incremental scan: The incremental scan is automatically executed by the vision unit: the workpiece point cloud volume is updated after each grab; when the current volume is reduced by more than 30% compared with the last scan, a rescan is triggered; New frame scan: After the material frame is in place, the sensor outputs a ready signal, and a special scan for that station is immediately triggered.

[0038] 2. For example Figure 2 As shown, the multi-station collaborative logic state machine control rules are as follows: Full material → In operation: Triggered when the robot first grabs workpiece 11 at the material frame station; In operation → Awaiting replacement: Meets any of the following conditions: the number of workpieces is less than 5% of the full material state, three consecutive grab failures, awaiting replacement → Full material: the material frame position sensor continuously outputs a ready signal for more than 2 seconds; Control flow: ① When the material frame station enters the awaiting replacement state, its grabbing command is frozen and an audible and visual alarm is triggered; ② The robot automatically switches to other stations to continue operation; ③ After manual frame replacement, the sensor ready signal triggers a state update and a new material frame scan.

[0039] 3. The continuous capture failure counter is automatically reset to zero after the state switches to "Pending Replacement". The specific implementation logic is as follows: When the state machine executes the "Working → Pending Replacement" switch, the PLC sends a reset command to the vision unit. After receiving the command, the vision unit immediately resets the failure count value of the station to 0. After the new material frame is scanned, the system starts counting the number of failures again from the initial state. Note: The reset operation takes precedence over the material replacement alarm signal output to ensure the atomicity of the state switch.

[0040] Examples and Validation

[0041] Example 1: Loading bearing rollers.

[0042] Workpiece specifications: Cylindrical roller, 15 mm in diameter and 20 mm in height. Parameter configuration: Gripping height: 50 mm from the bottom of the material frame station 11; Fault tolerance threshold: Triggering replacement after 3 consecutive gripping failures; Scanning density: 2 point cloud sampling points per square millimeter. Performance data: Average gripping cycle per piece: 1.2 seconds; Gripping capacity: 21,500 pieces in 8 hours of continuous operation; Equipment utilization rate: 100% (zero downtime for material changes).

[0043] Counter reset verification: Simulate 3 consecutive failed grabs → status switch to wait for replacement → counter reset to zero; the first grab success rate after the new material frame is in place reaches 100% (tested 50 times).

[0044] Volume calculation accuracy verification Actual number of workpieces Visual system calculated values error 1000 pieces 982 items -1.8% 500 pieces 510 items +2.0%

[0045] Example 2: Loading the gear blank.

[0046] Workpiece specifications: Module 2, 45 mm outer diameter alloy gear. Technical adjustments: Camera activated polarization filter mode to suppress metal reflection; fault tolerance threshold reduced to two consecutive failures.

[0047] Comparison data: System type Bearing roller gripping success rate Gear blank gripping success rate Traditional single-frame system 92.3% 85.1% This invention system 99.7% 98.9%

[0048] Industrial application prospects:

[0049] Automotive parts: batch feeding of bearing rollers and gear blanks; 3C electronics: disordered gripping of mobile phone casings and heat sinks; medical devices: flexible feeding of surgical forceps handles and joint bearings.

[0050] Finally, several points should be noted: First, in the description of this application, it should be noted that, unless otherwise specified and limited, the terms "installation," "connection," and "linkage" should be interpreted broadly, and can refer to mechanical or electrical connections, or internal connections between two components, or direct connections. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may change. Second, the accompanying drawings of the embodiments disclosed in this invention only involve structures relevant to the embodiments disclosed in this invention; other structures can refer to common designs. Where there is no conflict, the same embodiment and different embodiments of this invention can be combined with each other. Finally, the above descriptions are merely preferred embodiments of this invention and are not intended to limit this invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A multi-position camera positioning system, characterized in that, Its structure includes: a mobile vision positioning unit (1), a multi-station material frame carrying area, a robot execution unit and a control system; the multi-station material frame carrying area is provided with at least two independent material frame stations (11); the mobile vision positioning unit (1) moves above each material frame station (11) through a horizontal moving mechanism (2) and performs 3D vision positioning on the scattered workpieces in the material frame station (11) to generate real-time updated workpiece positioning data; the control system generates grasping instructions based on the workpiece positioning data and controls the robot execution unit to grasp workpieces in different material frame stations (11) across stations; wherein, when a certain material frame station (11) is in an empty frame replacement state, the robot execution unit continues to grasp work according to the pre-stored positioning data of the other material frame stations (11).

2. The multi-position camera positioning system according to claim 1, characterized in that, The horizontal moving mechanism (2) includes a support column (21), a linear guide rail (22), and a drive motor (23); the linear guide rail (22) is horizontally fixed on the support column (21) and spans the multi-station material frame bearing area; the linear guide rail (22) is provided with a sliding bracket (3); the mobile vision positioning unit (1) is mounted on the bracket (3) in a liftable manner; the drive motor (23) drives the bracket (3) to move horizontally along the linear guide rail (22) through a transmission component.

3. The multi-position camera positioning system according to claim 2, characterized in that, The drive motor (23) is a servo motor and is equipped with a position detection element. The transmission component is a precision ball screw. The servo motor, the precision ball screw and the position detection element constitute a closed-loop control system with a horizontal positioning accuracy of ≤0.1mm.

4. A multi-position camera positioning system according to claim 2 or 3, characterized in that, The mobile visual positioning unit (1) is a 3D camera.

5. A multi-position camera positioning system according to claim 1, characterized in that, The multi-station material frame carrying area is provided with three material frame stations (11), which are arranged linearly or in a circular array; each of the material frame stations (11) is equipped with a material frame positioning sensor; the material frame positioning sensor is used to detect whether the material frame is in an operable position.

6. The multi-position camera positioning system according to claim 1, characterized in that, The robot execution unit is a six-axis industrial robot with a vacuum suction cup or adaptive gripper installed at the end.

7. A multi-position camera positioning system according to claim 5, characterized in that, The mobile vision positioning unit (1) executes a dynamic scanning strategy: a) Initial scan: scan all material frame stations (11) in sequence to generate global workpiece coordinates; b) Incremental scan: when the change in the number of workpieces at a certain material frame station (11) exceeds the set value, a local re-scan is triggered; c) New frame scan: when the material frame positioning sensor detects that a new material frame is in place and its status is to be replaced, a special scan of the material frame station (11) is triggered.

8. A multi-position camera positioning system according to claim 7, characterized in that, The multi-station collaborative logic of the control system includes: (1) establishing a status identifier for each material frame station (11), the status identifier including full material status, working status and waiting-to-be-replaced status; (2) when the status of a material frame station (11) is switched to the waiting-to-be-replaced status: freezing the grabbing instruction of the material frame station (11) and activating the pre-stored positioning data of the other material frame stations (11); (3) after receiving the replacement signal of the material frame in place sensor, updating the status identifier and triggering the new frame scanning.

9. A multi-position camera positioning system according to claim 8, characterized in that, The switching rules for the status identifier include: switching from full status to working status: triggered when the material frame station (11) is grabbed for the first time; switching from working status to waiting-to-replace status: triggered when the visual scan detects that the number of workpieces is lower than the set threshold or the number of consecutive grabbing failures reaches the fault tolerance limit; switching from waiting-to-replace status to full status: triggered after receiving the material frame positioning sensor signal.

10. A multi-position camera positioning system according to claim 1, characterized in that, The system is applied to automated feeding scenarios for bearing rollers, gear blanks, and injection molded parts.

Images