Palletizers are crucial end-of-line equipment in modern automated production lines, capable of stacking and transporting goods. They can be integrated with various packaging machines, significantly improving production efficiency. “Identifying and gripping objects” is the primary and critical step for a palletizer to complete its tasks. This article will delve into the basic structure of palletizers, and provide a detailed analysis of their positioning and gripping technologies and applications.
Basic Structure of Palletizers
Palletizing robots typically consist of a robotic arm, gripper, sensors, and a control system. Among these, the robotic arm is responsible for the movement and precise positioning of goods; the gripper or suction cup acts as the execution mechanism, used for clamping or picking up items; sensors serve as the “eyes” and “touch” of the palletizing robot, responsible for perceiving the surrounding environment and the position and orientation of target objects; and the control system is the “brain” of the entire robot, receiving and processing sensor data, planning movement trajectories, and coordinating the collaborative work of various components.
Positioning and gripping are the core elements for a palletizing robot to complete palletizing tasks. During this process, the robot needs to obtain key information such as the position, orientation, and quantity of target items in real-time through various sensors, thereby precisely controlling the gripper to complete the gripping action. Sensors play a vital role in palletizing and can be divided into two main categories:
- Non-vision sensors: These include photoelectric sensors, proximity sensors, laser sensors, and force sensors, mainly used to detect the presence of objects, measure distances, determine positions, and sense force feedback during the gripping process.
- Vision sensors: These primarily refer to various industrial cameras. By capturing images of objects and utilizing complex image processing algorithms, they enable the identification, positioning, and orientation determination of objects.
Positioning and Gripping Technologies of Palletizers
Palletizer positioning and gripping technologies mainly include the following four types:
Non-vision Sensor Technology
- Photoelectric Sensor[wikipedia]: The Photoelectric sensor is the most basic identification method for determining whether an object is on a conveyor line and its location by emitting and receiving infrared light. By reflecting light or emitting light signals, the photoelectric sensor can estimate the distance between the object and the sensor for localization and tracking, ensuring high production line efficiency. It is often used in palletizing operations to detect whether the material on the conveyor belt reaches the preset gripping point, and to detect whether the pallet is in place or the palletizing height.
- Proximity Sensor: The sensor utilizes principles such as electromagnetic induction, capacitive sensing, or ultrasonic waves to detect the proximity of objects without direct contact. On palletizers, the proximity sensor is often used to detect whether the robotic arm has moved to the predetermined horizontal or vertical position, or to determine whether an object has entered the effective gripping range of the gripper.
- Laser Sensor: The laser sensor achieves highly accurate distance measurement and object contour scanning by emitting high-precision laser beams and measuring the time or intensity changes of their reflections. Some palletizers that need to handle randomly stacked materials use laser sensors to scan the three-dimensional contours of the material pile, obtaining the position and size information of each object, providing precise data support for subsequent gripping.
- Force Sensor: The force sensor is installed at the end of the robot arm or inside the gripper. When the gripper grabs the goods, the force sensor detects a change in force. When the force sensor detects a force value within a predetermined range, it indicates successful gripping. If the force value changes abnormally or does not stabilize, the gripping fails. The palletizer control system can control the gripping strength of the gripper by receiving the force value feedback, to avoid damaging the goods due to excessive gripping force or dropping the goods due to insufficient gripping force.
Vision Sensor Technology
- 2D Vision: 2D vision systems use industrial cameras to capture two-dimensional images of objects, and then use complex image processing algorithms (such as edge detection, shape matching, color recognition, template matching, etc.) to identify the position, shape, size, and orientation of objects. This technology is suitable for situations where the objects to be gripped are arranged relatively regularly on the conveyor belt without significant occlusion.
- 3D Vision: 3D vision systems go a step further by using 3D cameras (such as stereo vision cameras, structured light cameras, time-of-flight cameras, etc.) to obtain three-dimensional point cloud data of objects. Compared to 2D images, 3D data can describe the spatial information of objects more comprehensively and accurately. Even irregularly stacked or irregularly shaped objects with partial occlusion can be accurately identified and positioned. 3D vision systems typically combine artificial intelligence and deep learning algorithms, and through learning from a large amount of data, possess stronger robustness and recognition capabilities, enabling them to handle object recognition and positioning tasks in various complex scenarios.
- Vision Guidance: Whether it’s a 2D or 3D vision system, once it successfully identifies a target object, it transmits the object’s position, orientation, and other information to the palletizer’s control system in real-time. Based on this information, the control system precisely plans the movement trajectory of the robotic arm, guides the gripper to accurately move to the target position, and adjusts the gripping posture to finally complete the precise gripping action.
Pre-programming Technology
In many automated production lines, the specifications, shapes, and arrival positions of the objects to be palletized on the conveyor belt are relatively uniform and fixed. In this case, engineers can pre-program the detailed parameters of these objects into the palletizer’s control system and establish a corresponding database. When sensors installed on the conveyor belt (usually photoelectric sensors) detect that an object has reached the preset detection point, the palletizer can directly extract the gripping parameters of the object from the pre-stored database and perform gripping according to the predetermined program without complex real-time identification. The advantages of this method are high speed and low cost, but it has poor flexibility and is suitable for palletizing single-variety, fixed-specification materials.
Barcode/QR Code Scanning
Some boxes are labeled with a specific Barcode/QR Code, which is used to record the type of item, its specifications, date of manufacture, etc. The palletizer can be equipped with a corresponding scanner. The palletizer can be equipped with a corresponding scanner to scan and read the Barcode/QR Code information and combined with the preset palletizing rules to accurately control the robotic arms and grippers to complete the positioning and stacking of goods.
III. Applications
When palletizing items of consistent weight, size, and specifications, the most commonly used positioning and gripping method is pre-programming combined with sensors. This involves precisely pre-programming the palletizer’s movement trajectory and gripper actions, and using photoelectric sensors to detect the arrival signal of bags on the conveyor belt, triggering the palletizer to execute the preset gripping and placement actions. The pre-programmed action flow is very smooth and fast. Combined with sensor triggering, it can achieve higher palletizing speeds, and these are relatively mature and low-cost technologies, which can simplify the palletizer’s control system and reduce the difficulty of operation and maintenance.
2D/3D vision systems are mainly used when it is necessary to handle objects with inconsistent shapes, sizes, or colors, when objects are arranged disorderly on the conveyor belt or in a material pile, when quality inspection or object feature recognition is required, or when complex palletizing patterns are needed. They can handle more complex palletizing operations, and therefore the cost will also be higher. In addition, the performance of vision systems is easily affected by changes in ambient light, and the system integration and debugging are difficult, and the operation difficulty will also be greater.
Barcode/QR code scanning technology can be used when products need to be palletized by type, specification, production batch, etc. In practical applications, it is necessary to ensure that the barcode or QR code is printed clearly and to avoid environmental interference factors such as dust, water mist, and strong light to ensure the stability and reliability of the scanning system, control system, and integrated system. It is necessary to equip corresponding scanners, data acquisition terminals, and corresponding software and integrated systems, which are relatively expensive.