Intelligent robot logistics system based on multiple sensors and optimal control

Abstract

The invention discloses an intelligent robot logistics system based on multiple sensors and optimization control, and belongs to the technical field of intelligent logistics automation. Aiming at the problems of low path planning efficiency, insufficient motion stability and the like of a traditional logistics system, the system constructs an environment sensing module through a high-precision camera, a laser radar and an ultrasonic sensor, combines an improved A star algorithm, introduces a turning cost function to optimize the path length and the turning times, realizes efficient path planning, and improves the path planning efficiency. And motion parameters are dynamically adjusted by adopting a fuzzy-PID composite control algorithm, so that the stability under a complex working condition is improved. The method is suitable for warehousing, e-commerce distribution and industrial automation scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of logistics automation, and in particular to an intelligent robot logistics system based on multi-sensors and optimized control. Background Art

[0002] In the traditional logistics industry, handling work mostly relies on manual operation, which not only has high labor costs, but also has low efficiency and is prone to human errors. With the explosive growth of the e-commerce industry and the continuous expansion of the scale of warehousing logistics, the traditional logistics model has been difficult to meet the needs of modern efficient logistics. In recent years, with the rapid development of the logistics industry and the continuous progress of technologies such as artificial intelligence and the Internet of Things, intelligent robot logistics systems have gradually become an important development direction in the modern logistics field. As the core component of logistics automation, intelligent robot logistics systems can not only effectively improve the efficiency of logistics operations, but also reduce labor costs, reduce human errors, and improve the overall quality of logistics services.

[0003] Although logistics systems based on robots are gradually emerging, current intelligent robot logistics systems still face many challenges in practical applications, especially in aspects such as intelligent perception systems, path planning, motion control, and environmental adaptability. For example, the perception ability of a single sensor is limited and it is difficult to cope with complex and changeable logistics environments; traditional path planning algorithms, such as the A* algorithm, although able to find the shortest path in complex environments, do not fully consider the number of turns, resulting in redundant paths and increased energy consumption, and it is difficult to achieve true shortest path planning. Frequent turning operations not only reduce the handling efficiency, but also increase the wear and energy consumption of the equipment; in addition, in the motion control link, due to its own limitations, the conventional PID control method is difficult to adapt to complex environments. In the face of complex working conditions, such as ground bumps and load changes, it is unable to adjust control parameters in a timely manner according to the real-time changing environment and equipment status, resulting in insufficient motion stability of the robot, which is prone to cause damage to goods and affects the overall efficiency of the logistics system.

[0004] In summary, developing an intelligent robot logistics system that can effectively optimize path planning and significantly enhance motion stability is of crucial practical significance for promoting the modernization development of the logistics industry. Summary of the Invention

[0005] The purpose of the present invention is to overcome the above-mentioned defects in the background art, and provide an intelligent robot logistics system based on multi-sensors and optimized control. By using multiple sensors and adopting an improved A* algorithm and a fuzzy-PID composite control algorithm, it has extremely high versatility and adaptability, can be widely applied to multiple fields such as warehousing logistics, e-commerce distribution, and factory production, provides strong support for the intelligent upgrading of these industries, and has broad market application prospects and commercial value.

[0006] The present invention adopts the following technical solutions to solve the above technical problems:

[0007] An intelligent robot logistics system based on multi-sensors and optimized control, the logistics system includes: an intelligent perception system, an optimized path planning module, and an advanced motion control module; the intelligent perception system is jointly constructed by a diversified sensor composed of an ultrasonic sensor and a lidar equipped on the intelligent robot and a high-definition camera, the optimized path planning module adopts an improved A* algorithm, and the advanced motion control module adopts a fuzzy-PID composite control algorithm.

[0008] Further, the working steps of the intelligent robot logistics system are as follows:

[0009] Step S1: Set the initial position of the intelligent robot and prepare to start working;

[0010] Step S2: Activate the sensor to detect the current position and drive along a predetermined route;

[0011] Step S3: The high-definition camera identifies the QR code to achieve precise positioning;

[0012] Step S4: The intelligent robot moves to the initial position of the goods and prepares for the next operation;

[0013] Step S5: The camera works again to determine the status of the goods;

[0014] Step S6: The intelligent robot uses its clamping device to grab the goods and starts the handling process. During the handling process, the system performs path planning to determine the best route from the current position to the target position, and uses the fuzzy-PID composite control algorithm to control the movement of the robot to ensure that the goods are transported to the designated position smoothly and accurately;

[0015] Step S7: After completing one handling, the intelligent robot returns to the initial position or waits for the next instruction to prepare for the next handling task.

[0016] Further, the laser beam emitted by the lidar travels to an object at a distance of d and reflects back within time t. According to the constant speed of light c, the distance between the object and the lidar is calculated using the formula:

[0017]

[0018] By continuously changing the emission direction of the laser beam, a series of distance data in different azimuths are obtained, and then these data are used to construct a three-dimensional map of the surrounding environment to support the robot's cruise.

[0019] Further, the ultrasonic sensor uses the propagation speed v of ultrasonic waves in the air to measure the distance and keenly sense obstacles at close range. When the sensor emits an ultrasonic wave signal and receives the reflected signal after time t, the distance L to the obstacle can be calculated according to the formula:

[0020]

[0021] Further, the high-definition camera adopts advanced image acquisition technology to clearly capture images of the surrounding environment and goods; in the image recognition link, the convolutional neural network in the deep learning algorithm is used. The basic structure of the convolutional neural network includes: convolutional layer, pooling layer, and fully connected layer; in the convolutional layer, the convolution operation is performed between the convolution kernel and the image to extract the features of the image. The convolution operation is expressed as:

[0022] O(x,y)=∑ m,n I(x+m,y+n)×K(m,n)

[0023] where O(x,y) is the value of the output feature map at position (x,y), I is the input image, and K is the convolution kernel;

[0024] After continuously extracting and compressing features through multiple convolutional layers and pooling layers, the obtained feature map is input into the fully connected layer for classification and recognition; by constructing a photo training set and performing data augmentation in a 7:3 ratio, the model learns the unique feature patterns of various goods, so as to accurately identify the position, shape, and size of the goods, and determine the type and current state of the goods accordingly.

[0025] Further, the improved A* algorithm utilizes the accurate map model constructed from the environmental information obtained by the sensor. The map is divided into 100*100 grids. The algorithm starts from the starting point, calculates the comprehensive cost F(n) of each node, selects the node with the minimum cost as the next expansion node, and gradually searches for the target point. During the search process, the heuristic function h(n) is used to guide the search direction and accelerate the search speed.

[0026] Further, the improved A* algorithm adds the consideration of the number of turns on the basis of the traditional A* algorithm, introduces the turning cost function t(n), and searches for the optimal path by calculating the improved evaluation function F(n) of the node:

[0027] F(n)=g(n)+h(n)+ω×t(n)

[0028] where g(n) represents the actual cost from the starting point to the current node n, that is, the length of the path already traveled, h(n) is the estimated cost from the current node n to the target point, calculated using the heuristic function, ω is the turning cost weight coefficient, and t(n) represents the cumulative value of the number of turns from the starting point to the current node n;

[0029] Furthermore, the fuzzy-PID composite control algorithm includes a fuzzy control part and a PID control part. The fuzzy control part performs fuzzy processing on the speed deviation and its change rate of the robot, and conducts inference operations according to the fuzzy rule base to obtain a fuzzy output. The PID control part adjusts the control parameters of the motor according to the output result of the fuzzy control.

[0030] Furthermore, the steps of the fuzzy control part of the fuzzy-PID composite control algorithm are as follows:

[0031] Step S6-1: Calculate the speed deviation e and its change rate ec of the robot:

[0032] e = v sct - v actual

[0033]

[0034] where v sct is the set speed, v actual is the actual speed, T is the sampling period, e(k) and e(k - 1) are the speed deviations at the current and previous moments respectively;

[0035] Step S6-2: Map e and ec to the corresponding fuzzy universes of discourse, and describe them with fuzzy linguistic variables "negative large", "negative medium", "negative", "zero", "positive small", "positive medium", "positive" as fuzzy inputs;

[0036] Step S6-3: Conduct inference operations according to the pre-set fuzzy rule base. The fuzzy rules are presented in the form of "if-then", and the fuzzy output is obtained by performing fuzzy inference operations on the fuzzy inputs;

[0037] Step S6-4: Perform defuzzification processing, and use the centroid method to convert the fuzzy quantity into an accurate adjustment amount. The formula is:

[0038]

[0039] where, u 0 is the accurate adjustment amount after defuzzification, u i is an element in the fuzzy universe of discourse, and μ(u i ) is its corresponding membership degree.

[0040] Furthermore, for the PID control part of the fuzzy-PID composite control algorithm, the three control parameters of the motor are accurately adjusted according to the output result of the fuzzy control. The calculation formula for the control quantity u(t) of the PID control algorithm is:

[0041]

[0042] where K p is the proportionality coefficient, K i is the integral coefficient, K d is the differential coefficient, and e(t) is the system deviation, that is, the adjustment amount of the fuzzy control output is used to online adjust the three parameters K p , K i , K d to help adapt to different motion states and environmental changes.

[0043] Compared with the prior art, the present invention adopting the above technical solution has the following beneficial effects:

[0044] (1) An intelligent robot logistics system based on multi-sensors and optimized control provided by the present invention supports multi-sensor expansion and algorithm modular transplantation, can be adapted to devices such as AGVs and unmanned forklifts, and has strong versatility;

[0045] (2) An intelligent robot logistics system based on multi-sensors and optimized control provided by the present invention adopts an improved A* algorithm to comprehensively optimize the path length and the number of turns. During the path finding process, the algorithm always selects the node with the smallest F(n) value for expansion, gradually searches for the target point, and will try to avoid unnecessary turns, thereby effectively reducing the handling time and energy consumption, and the handling efficiency is increased by 25%-40%;

[0046] (3) An intelligent robot logistics system based on multi-sensors and optimized control provided by the present invention adopts fuzzy-PID composite control to adaptively adjust parameters, and the bumpiness is reduced to below 0.5g (international standard ≤ 1.2g), improving the smoothness of the robot movement. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is an introduction diagram of an intelligent robot logistics system based on multi-sensors and optimized control of the present invention.

[0048] Figure 2 is a flowchart of the operation of an intelligent robot logistics system based on multi-sensors and optimized control of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0049] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings of the specification. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0050] An intelligent robot logistics system based on multi-sensors and optimized control, the logistics system includes: an intelligent perception system, an optimized path planning module, and an advanced motion control module, as Figure 1 shown; the intelligent perception system is jointly constructed by a diversified sensor composed of an ultrasonic sensor and a lidar equipped on the intelligent robot and a high-definition camera, the optimized path planning module adopts an improved A* algorithm, and the advanced motion control module adopts a fuzzy-PID composite control algorithm; the advanced motion control module adopts a fuzzy-PID composite control algorithm; the hardware equipment of the intelligent robot selects an STM32F4 series controller, a 360° lidar with a detection radius of 10m, a 2-million-pixel camera, and a DC brushless motor. The software develops path planning and motion control programs based on the ROS system, and the map grid resolution is set to 5cm×5cm.

[0051] Further, as Figure 2 shown, the working steps of the intelligent robot logistics system are as follows:

[0052] Step S1: Set the initial position of the intelligent robot and prepare to start working;

[0053] Step S2: Activate the sensor to detect the current position and drive along a predetermined route;

[0054] Step S3: The high-definition camera recognizes the QR code to achieve precise positioning;

[0055] Step S4: The intelligent robot moves to the initial position of the goods and prepares for the next operation;

[0056] Step S5: The camera works again to determine the status of the goods;

[0057] Step S6: The intelligent robot uses its clamping device to grab the goods and starts the handling process. During the handling process, the system performs path planning to determine the best route from the current position to the target position, and uses a fuzzy-PID composite control algorithm to control the movement of the robot to ensure that the goods are smoothly and accurately transported to the designated position;

[0058] Step S7: After completing one handling, the intelligent robot returns to the initial position or waits for the next instruction to prepare for the next handling task.

[0059] Further, the laser beam emitted by the lidar travels to an object at a distance of d and reflects back within time t. According to the constant speed of light c, the distance between the object and the lidar is calculated using the formula:

[0060]

[0061] By continuously changing the emission direction of the laser beam, a series of distance data in different orientations are obtained, and then these data are used to construct a three-dimensional map of the surrounding environment to support the robot's cruising. For example, in the Cartesian coordinate system, through multiple distance data (d 1 , θ 1 ), (d 2 , θ 2 ), …, it can be converted into coordinate points (x i , y i , z i ), where x i = d i × cos(θ i ), y i = d i × sin(θ i ), and z i is determined according to the actual measurement scenario. These coordinate points together constitute a three-dimensional model of the environment, providing basic support for the robot's navigation in a complex environment.

[0062] Furthermore, the ultrasonic sensor uses the propagation speed v of ultrasonic waves in the air to measure the distance and keenly sense the obstacles in the near distance. When the sensor emits an ultrasonic wave signal and receives the reflected signal after time t, the distance L to the obstacle can be calculated according to the formula:

[0063]

[0064] Furthermore, the high-definition camera adopts advanced image acquisition technology to clearly capture the images of the surrounding environment and goods; in the image recognition link, the convolutional neural network (CNN) in the deep learning algorithm is used. The basic structure of the CNN includes: convolutional layer, pooling layer, and fully connected layer; in the convolutional layer, through the convolution operation of the convolution kernel and the image, the features of the image are extracted, and the convolution operation is expressed as:

[0065] O(x, y) = ∑ m,n I(x + m, y + n) × K(m, n)

[0066] where O(x, y) is the value of the output feature map at the position (x, y), I is the input image, and K is the convolution kernel;

[0067] After continuously extracting and compressing the features through multiple convolutional layers and pooling layers, the obtained feature map is input into the fully connected layer for classification and recognition; by constructing a training set containing three thousand data photos and performing data augmentation in a ratio of 7:3, the model can learn the unique feature patterns of various goods, which enables the model to accurately identify the position, shape, and size of the goods, and determine the type and current state of the goods accordingly.

[0068] Furthermore, the improved A* algorithm utilizes the environmental information obtained by sensors to construct an accurate map model, divides the map into multiple grid nodes, starts from the starting point, calculates the comprehensive cost F(n) of each node, selects the node with the minimum cost as the next expansion node, and gradually searches for the target point. During the search process, the heuristic function h(n) is used to guide the search direction and accelerate the search speed.

[0069] Furthermore, the improved A* algorithm takes into account the number of turns on the basis of the traditional A* algorithm, introduces the turning cost function t(n), and finds the optimal path by calculating the improved evaluation function F(n) of the node:

[0070] F(n) = g(n) + h(n) + ω × t(n)

[0071] where g(n) represents the actual cost from the starting point to the current node n, that is, the length of the path already traveled, h(n) is the estimated cost from the current node n to the target point, calculated using the heuristic function, ω is the turning cost weight coefficient, and t(n) represents the cumulative value of the number of turns from the starting point to the current node n;

[0072] Furthermore, the fuzzy-PID composite control algorithm includes a fuzzy control part and a PID control part. The fuzzy control part performs fuzzy processing on the speed deviation and its change rate of the robot, and performs reasoning operations according to the fuzzy rule base to obtain a fuzzy output. The PID control part adjusts the control parameters of the motor according to the output result of the fuzzy control.

[0073] Furthermore, the steps of the fuzzy control part of the fuzzy-PID composite control algorithm are as follows:

[0074] Step S6-1: Calculate the speed deviation e and its change rate ec of the robot:

[0075] e = v sct - v actual

[0076]

[0077] where v sct is the set speed, v actual is the actual speed, T is the sampling period, and e(k) and e(k - 1) are the speed deviations at the current and previous moments respectively;

[0078] Step S6-2: Map e and ec to the corresponding fuzzy universes, and describe them with fuzzy language variables "negative large", "negative medium", "negative small", "zero", "positive small", "positive medium", "positive large" as fuzzy inputs;

[0079] Step S6-3: Conduct inference operations according to a pre-set fuzzy rule base. The fuzzy rules are presented in the form of "if-then". By performing fuzzy inference operations on the fuzzy inputs, fuzzy outputs are obtained;

[0080] Step S6-4: Perform defuzzification processing. Use the centroid method to convert the fuzzy quantity into an accurate adjustment quantity. The formula is:

[0081]

[0082] where, u 0 is the accurate adjustment quantity after defuzzification, u i is an element in the fuzzy domain, and μ(u i ) is its corresponding membership degree.

[0083] Furthermore, for the PID control part of the fuzzy-PID composite control algorithm, the three control parameters of the motor are accurately adjusted according to the output result of the fuzzy control. The calculation formula for the control quantity u(t) of the PID control algorithm is:

[0084]

[0085] where K p is the proportionality coefficient, K i is the integral coefficient, K d is the differential coefficient, e(t) is the system deviation, that is, the adjustment quantity output by the fuzzy control is used to online adjust the three parameters K p , K i , K d to help adapt to different motion states and environmental changes.

[0086] In addition, the above has introduced in detail an intelligent robotic logistics system based on multi-sensors and optimized control provided by the embodiments of the present invention. Specific examples should have been used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. An intelligent robot logistics system based on multi-sensors and optimized control, characterized in that: The logistics system includes: an intelligent perception system, an optimized path planning module, and an advanced motion control module; the intelligent perception system is constructed by a diversified sensor consisting of ultrasonic sensors and laser radars equipped by intelligent robots and high-definition cameras, the optimized path planning module adopts an improved A-star algorithm, and the advanced motion control module adopts a fuzzy-PID composite control algorithm.

2. According to claim 1, the intelligent robot logistics system based on multi-sensor and optimization control is characterized in that: The working steps of the intelligent robot logistics system are as follows: Step S1: The initial position of the intelligent robot is set and it is ready to start working; Step S2: activating sensors to detect the current location and travel along the predetermined route; Step S3: The high-definition camera recognizes the QR code to achieve precise positioning; Step S4: The intelligent robot moves to the initial position of the goods and prepares for the next operation; Step S5: The camera works again to determine the status of the goods; Step S6: The intelligent robot uses its gripping device to grab the goods and start the handling process. During the handling process, the system performs path planning to determine the best route from the current position to the target position, and uses the fuzzy-PID composite control algorithm to control the movement of the robot to ensure that the goods are delivered to the designated location smoothly and accurately; Step S7: After completing one transport task, the intelligent robot returns to the initial position or waits for the next instruction to prepare for the next transport task.

3. According to claim 1, the intelligent robot logistics system based on multi-sensor and optimization control is characterized in that: The laser beam emitted by the laser radar propagates to an object at a distance d within time t and is reflected back. According to the constant speed of light c, the distance between the object and the laser radar is calculated using the formula: By continuously changing the emission direction of the laser beam, a series of distance data in different directions is obtained, and then these data are used to build a three-dimensional map of the surrounding environment to support robot cruising.

4. The intelligent robot logistics system based on multi-sensor and optimized control according to claim 1 is characterized in that: The ultrasonic sensor uses the propagation speed v of ultrasonic waves in the air to measure distance and sense close obstacles. When the sensor transmits an ultrasonic signal and receives the reflected signal after a period of time t, the distance L to the obstacle can be calculated according to the formula:

5. The intelligent robot logistics system based on multi-sensors and optimized control according to claim 1 is characterized in that: The high-definition camera adopts advanced image acquisition technology to clearly capture images of the surrounding environment and cargo; In the image recognition process, the convolutional neural network in the deep learning algorithm is used. The basic structure of the convolutional neural network includes: convolution layer, pooling layer and fully connected layer. In the convolution layer, the convolution kernel is convolved with the image to extract the features of the image. The convolution operation is expressed as: O(x,y)=∑ m,n I(x+m,y+n)×K(m,n) Among them, O(x,y) is the value of the output feature map at position (x,y), I is the input image, and K is the convolution kernel; After multiple convolutional and pooling layers continuously extract and compress features, the obtained feature maps are input into the fully connected layer for classification and recognition. By constructing a photo training set and using a 7:3 ratio for data enhancement, the model learns the unique feature patterns of various types of goods, thereby accurately identifying the location, shape, and size of the goods, and determining the type of goods and their current status based on this.

6. The intelligent robot logistics system based on multi-sensor and optimized control according to claim 1 is characterized in that: The improved A-star algorithm uses the accurate map model constructed by the environmental information obtained by the sensor to divide the map into 100*100 grids. The algorithm starts from the starting point, calculates the comprehensive cost F(n) of each node, selects the node with the smallest cost as the next expansion node, and gradually searches to the target point. During the search process, the heuristic function h(n) is used to guide the search direction and speed up the search.

7. The intelligent robot logistics system based on multi-sensors and optimized control according to claim 1 is characterized in that: The improved A-star algorithm adds the consideration of the number of turns on the basis of the traditional A-star algorithm, introduces the turning cost function t(n), and finds the optimal path by calculating the improved evaluation function F(n) of the node: F(n)=g(n)+h(n)+ω×t(n) Among them, g(n) represents the actual cost from the starting point to the current node n, that is, the length of the path that has been traveled, h(n) is the estimated cost from the current node n to the target point, calculated using the heuristic function, ω is the turning cost weight coefficient, and t(n) represents the cumulative number of turns from the starting point to the current node n.

8. The intelligent robot logistics system based on multi-sensors and optimized control according to claim 1 is characterized in that: The fuzzy-PID composite control algorithm includes a fuzzy control part and a PID control part. The fuzzy control part performs fuzzy processing on the speed deviation and its change rate of the robot, and performs inference operation according to the fuzzy rule base to obtain a fuzzy output. The PID control part adjusts the control parameters of the motor according to the output result of the fuzzy control.

9. The intelligent robot logistics system based on multi-sensors and optimized control according to claim 1 is characterized in that: The fuzzy control part steps of the fuzzy-PID composite control algorithm are: Step S6-1: Calculate the robot's velocity deviation e and its rate of change ec: e=v sct -v actual where v sct is the set speed, v actual is the actual speed, T is the sampling period, e(k) and e(k-1) are the speed deviations at the current and previous moments respectively; Step S6-2: Map e and ec to the corresponding fuzzy domain, and use fuzzy language variables "negative large", "negative medium", "negative", "zero", "positive small", "positive medium", and "positive" to describe them as fuzzy input; Step S6-3: Perform reasoning operations according to a preset fuzzy rule base. The fuzzy rules are presented in the form of "if-then". The fuzzy input is subjected to fuzzy reasoning operations to obtain a fuzzy output; Step S6-4: Perform defuzzification processing and use the centroid method to convert the fuzzy amount into an accurate adjustment amount. The formula is: Among them, u0 is the precise adjustment amount after defuzzification, u i is an element in the fuzzy domain, μ(u i ) is its corresponding membership degree.

10. The intelligent robot logistics system based on multi-sensors and optimized control according to claim 1, characterized in that: The PID control part of the fuzzy-PID composite control algorithm accurately adjusts the three control parameters of the motor according to the output result of the fuzzy control. The control quantity u(t) of the PID control algorithm is calculated as follows: Where K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient, e(t) is the system deviation, that is, the adjustment amount of the fuzzy control output is used to adjust the three parameters K of PID online. p , K i , K d .

Images