Abstract: This paper introduces the overall scheme, structural design, and hydraulic transmission design of industrial robot teaching equipment for logistics warehouse pickup and delivery. It provides conditions for practical training, creates a good environment for students to develop new thinking, and cultivates students' skills in correctly designing PLC control systems and adjusting and troubleshooting hydraulic transmission systems.
Keywords: x-9 robot; degrees of freedom; picking up goods; delivery
To meet the requirements of cultivating highly skilled application-oriented talents and to develop students' comprehensive practical and technical application abilities, we have designed and manufactured an industrial robot for picking and delivering goods in a logistics warehouse (hereinafter referred to as an industrial robot) to provide students with training in design, programming, adjustment, and troubleshooting skills.
1. Performance and Structure of Industrial Robots
1.1 Performance of Industrial Robots
This design adopts a cylindrical coordinate system (as shown in Figure 1), with four degrees of freedom and five motions. The motions and main parameters have been determined through debugging as follows.
The arm's extension and retraction (x-direction movement degree of freedom); stroke 300 mm, speed 3000 mm/min.
Industrial robot walking (Y-direction movement degree of freedom): travel distance 4000 nm, walking speed 400 mm/min. Robot motion degree of freedom arm lifting (Z-direction movement degree of freedom): 2000 mm/min.
Arm rotation (Z-direction rotational degree of freedom) limit adjustment. Rotation angle is 90° . Adjustable stop limit adjustment.
Claw opening and clamping: The distance between the two claws is 230mm when open and 200mm when clamped, with a clamping force of not less than 1kN. Adjustable, with a maximum gripping weight of 300N.
The positioning accuracy of the goods during pickup and delivery is 5mm.
1.2 Structure of Industrial Robots
Based on the parameter requirements of the overall plan and the actual manufacturing capabilities of Wuxi Vocational and Technical College, the structural design of each component and the whole machine was determined.
1.2.1 Walking Mechanism
The traveling mechanism employs a four-wheeled, rail-mounted system. As shown in Figure 2, the traveling mechanism is powered by a servo motor that transmits power to a cycloidal reducer via a synchronous pulley and belt. The reducer's output shaft then transmits power to the drive wheel via the synchronous pulley and belt. The traveling components are mounted on the traveling base plate.
1.2.2 Organism
The machine body adopts a structure in which the lower base plate and the upper top plate are connected by double guide columns. The upper top plate is equipped with the arm lifting servo motor, and the lower base plate is connected to the body rotation shaft; the columns are the guide columns of the arm lifting components.
1.2.3 Arm extension
The boom's extension and retraction are achieved via hydraulic transmission using a double-acting, single-extend rod hydraulic cylinder; it employs a double-guide column, double-guide structure. The boom's extension and retraction cylinder is mounted on the moving plate of the boom lifting component.
1.2.4 Lifting and lowering of the boom
The arm's lifting mechanism is driven by a ball screw and nut system. A DC servo motor is mounted on the top plate of the machine body. The servo motor's output shaft is connected to the ball screw. The ball screw nut is mounted on the moving plate of the arm lifting mechanism. The ball screw is suspended, and gravity causes it to be under tensile stress.
1.2.5 Rotation of the arm
The arm's rotation is actually the rotation of the machine body. It is driven by two single-acting hydraulic cylinders, which in turn drive the shaft via a chain and sprocket mechanism. The rotation speed is adjusted by a throttle valve. The shaft's sliding bearing housing is connected to the mounting base plate. To withstand axial loads, a flat thrust ball bearing is installed at both the upper and lower ends of the shaft's sliding bearing.
1.2.6 Hand-held Meatball Mechanism
The gripper adopts a double-pivot linkage and lever structure. The opening and clamping of the gripper is achieved by hydraulic transmission of a double-acting single-extension hydraulic cylinder (as shown in Figure 3).
2. Control method and action cycle
2.1 Control Method
The system is an open-loop automatic control system. Position detection uses a contactless switch-type position sensor for sequence control and accurate positioning. The control method employs PLC programmable automatic control. Goods are retrieved from the platform and delivered to any location on the shelf, or retrieved from any location on the shelf and delivered to the platform (as shown in Figures 4 and 5). Alternatively, retrieval and delivery can be automatically controlled by computer programming, or manually or semi-automatically.
2.2 Action Cycle
When delivering goods: (goods are placed on the platform) the arm extends to the position, one hand grips tightly, one arm slightly raises, one arm retracts, and one walks (direction) to the position. One arm rotates 90 degrees (the arm rotates around the z-axis). One arm rises (in the z-direction) and moves to the position. One arm extends (in the z-direction) and moves to the position. One hand grip opens (goods are placed in the position). One arm retracts, one arm lowers, one arm reverses 90 degrees, and one walks to the front of the platform (waiting for new instructions or continuously delivering goods to the next position).
When picking up goods: The robot is in its original position in front of the platform. When it receives an instruction to pick up goods from a storage location, the robot walks (in the direction of movement) to the storage location, turns 90 degrees with one arm, raises (in the direction of movement) to the storage location height, extends (in the direction of movement) to the storage location, clamps the goods with one gripper, slightly raises one arm, retracts one arm, lowers to the storage location, and turns 90 degrees in the opposite direction. When it walks to the platform (in position), one arm extends to the storage location, one gripper opens (to put down the goods), and one arm retracts (waiting for new instructions or continuous picking up of goods in front of the platform).
3. Driving method
The arm lifting and walking are driven by DC servo motors. The arm extension and retraction, gripper opening and clamping, and arm rotation are driven by hydraulic transmission.
4. Hydraulic Transmission System Design
The extension and retraction of the arm, the rotation of the arm (body), and the opening and clamping of the gripper are all hydraulically driven. Based on the estimation of the clamping force, the extension and retraction torque of the arm, and the rotation torque of the arm, the diameters of each hydraulic cylinder and piston rod, the stroke of each hydraulic cylinder, and the pressure adjustment values of the hydraulic transmission system are determined. The principle of the hydraulic system is shown in Figure 6.
Figure 6 Working principle of hydraulic system
5. Main Practical Training Facilities
The training equipment for industrial robots used in logistics warehouse pickup and delivery mainly focuses on electrical control and hydraulic transmission training.
5.1 Electrical and Automatic Control Training Project
(1) Practical training on the basic principles of automatic control.
(2) Students design their own PLC program for practical training.
(3) On-machine debugging and operation training of PLC program control.
(4) Computer programming and online training.
5.2 Training Project on Debugging and Fault Analysis of Hydraulic Transmission System
(1) Basic practical training on the working principle and action cycle of hydraulic transmission.
(2) Debugging and training of hydraulic system working pressure, actuator movement speed, etc.
(3) Practical training on comprehensive electromechanical-hydraulic fault analysis and debugging, including hydraulic system oil pressure, flow rate, noise, action cycle, robot operation faults, etc.
The industrial robot teaching and training equipment for picking and delivering goods in logistics warehouses has undergone two years of development. Therefore, it is necessary to establish a mapping relationship between the fitness function and the objective function:
f(X)=KF(X) (6)
In the formula, K is a suitable positive constant to ensure that f(x) is always greater than 0.
2.3 Crossover
Crossover is a major operation in genetic algorithms. Crossover is actually an operation between the code strings of two individuals. The operation rule is to exchange the substrings between the two strings. The result is to exchange some genetic information between the two individuals, as shown in Figure 1.
In this algorithm, all individuals participating in the crossover operation are selected using the Fitness Proportional Mode 1, also known as the roulette wheel. The pairs of these individuals (forming parents) are determined randomly, and the strings involved in the crossover operation and their crossover positions are also randomly determined. The two new individuals (called offspring) generated from each pair of parent crossovers will serve as candidates for the next generation.
2.4 Mutation
Mutation is also an important operation in genetic algorithms. Mutation is essentially an independent operation performed by a single individual. The rule is to switch between 0s and 1s at a specific position in the individual's string, adding new information to the population. In the example presented in this paper, the individuals participating in mutation are selected with a small probability, and the mutation position is randomly determined. After mutation, a new individual with new genes is produced, and this new individual will also serve as a candidate for the next generation.
2.5 Algorithm Flowchart
The algorithm flow is shown in Figure 2, and it is implemented in the Matlab environment.
3. Calculation Results
Figure 3 shows the plastic part, made of ABS. In the algorithm, the population size is set to 0, the crossover probability is 0.5, the mutation probability is 0.001, and the maximum number of generations is 100. The weight coefficients A and B are each set to 0 and 5. The calculation results are shown in Figure 4.
Figure 4 shows that as the number of generations increases, the maximum fitness in each generation of the population tends towards "good". The data analysis also indicates that the fitness of the entire population also changes towards "good". The optimization results obtained from the general example are: injection temperature 279.99℃; mold temperature 31.6717℃; maximum injection pressure 160.37 MPa; maximum injection rate 696.27 cm³/s.
4. Conclusion
By leveraging existing CAE modules and employing genetic algorithms to optimize multi-parameter process conditions, the accuracy of which meets production process requirements can quickly provide accurate parameters for the selection of injection molding machines and the formulation of injection processes. Especially for large plastic parts, it can greatly reduce process adjustments during the trial molding stage, providing a reliable guarantee for the production of high-tech and high-quality parts.
