CN117474422B - Intelligent hillside orchard transportation system - Google Patents
Intelligent hillside orchard transportation system Download PDFInfo
- Publication number
- CN117474422B CN117474422B CN202311275155.1A CN202311275155A CN117474422B CN 117474422 B CN117474422 B CN 117474422B CN 202311275155 A CN202311275155 A CN 202311275155A CN 117474422 B CN117474422 B CN 117474422B
- Authority
- CN
- China
- Prior art keywords
- task
- conveyor
- real
- time
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000002420 orchard Substances 0.000 title claims abstract description 12
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 42
- 238000000034 method Methods 0.000 claims abstract description 12
- 230000007613 environmental effect Effects 0.000 claims abstract description 7
- 238000012544 monitoring process Methods 0.000 claims description 39
- 241000257303 Hymenoptera Species 0.000 claims description 14
- 238000002922 simulated annealing Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 11
- 239000003016 pheromone Substances 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 9
- 238000004088 simulation Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 230000001174 ascending effect Effects 0.000 claims description 6
- 238000001816 cooling Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 5
- 229910000831 Steel Inorganic materials 0.000 claims description 4
- 230000003068 static effect Effects 0.000 claims description 4
- 239000010959 steel Substances 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 2
- 238000012163 sequencing technique Methods 0.000 claims description 2
- 238000001914 filtration Methods 0.000 abstract description 5
- 238000000605 extraction Methods 0.000 abstract description 3
- 230000004888 barrier function Effects 0.000 abstract description 2
- 230000011218 segmentation Effects 0.000 abstract description 2
- 230000004927 fusion Effects 0.000 abstract 1
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 206010000117 Abnormal behaviour Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000002547 anomalous effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000001351 cycling effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000000725 suspension Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
- G06K17/0022—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device
- G06K17/0029—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device the arrangement being specially adapted for wireless interrogation of grouped or bundled articles tagged with wireless record carriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
Abstract
The invention relates to an intelligent hillside orchard transportation system. The invention adopts the fusion of two modes to realize the high-precision positioning of the conveyor. First, the system calculates mileage in real time by recording the revolution count of the conveyor drive wheels to infer the relative position of the conveyor on the track. The absolute position of the conveyor on the track can be determined by reading the RFID tags on the track with an RFID reader. By a combination of these two methods, the exact position of all on-track conveyors can be determined at low cost and with high accuracy. In addition, the system can calculate the distance between each conveyor in real time, and ensure that the obstacle avoidance strategy is effectively implemented. The conveyor is equipped with laser radar for acquire real-time environmental information, and cooperate the algorithm to realize data filtering, ground segmentation, barrier clustering, feature extraction and obstacle avoidance decision. The invention can realize safe obstacle avoidance and high-efficiency positioning, further realize task allocation and scheduling of multiple conveyors and improve the safety and the working efficiency of the conveyors.
Description
Technical Field
The invention relates to the field of agricultural machinery transportation scheduling, in particular to an intelligent hillside orchard transportation system.
Background
Currently in the prior art, some systems distribute scheduling work for tasks, relying on specific target scheduling rules to select vehicles that perform the commodity transportation tasks. This approach requires the generation of target scheduling rules based on the order parameters, logical relationships, and parameter thresholds corresponding to each sample rule, thereby limiting its applicability. On the other hand, some systems work for vehicle condition monitoring and positioning, employing various acquisition modules including load sensors, current sensors, speed sensors, voltage sensors, angle sensors, GPS positioning modules and cameras, but lack the provision of associated obstacle avoidance hardware, which makes them subject to collision risk during transportation. Moreover, the GPS positioning module utilized cannot achieve accurate positioning in tree-shading environments in mountainous areas.
Disclosure of Invention
The technical problem to be solved by the invention is to provide an intelligent hillside orchard transportation system aiming at the problems and requirements.
In order to solve the technical problems, the invention adopts the following technical scheme:
the intelligent hillside orchard transportation system comprises a dispatching system, a monitoring system and transportation machines, wherein the dispatching system is used for determining a feasible task distribution scheme of the transportation machines by using a simulated annealing algorithm according to the positions, the number and the task requirements of loading and unloading points on an annular single track, screening the feasible scheme by using an ant colony algorithm to obtain an optimal task distribution scheme, and distributing tasks to each transportation machine according to the optimal task distribution scheme;
the conveyor is used for loading and unloading at a specified loading and unloading point according to an allocated task, a control system, a current sensor, a voltage sensor, a load sensor, a two-axis angle sensor, a proximity switch sensor, an RFID reader and a 360-degree laser radar are arranged on the conveyor, and the current sensor and the voltage sensor are used for detecting real-time current and real-time voltage of a motor of the conveyor and uploading data to a monitoring system; the load sensor is used for detecting the real-time load capacity of the container and uploading data to the monitoring system; the two-axis angle sensor is used for detecting the left-right swing angle and the real-time upward-downward slope angle of the conveyor and uploading data to the monitoring system; the proximity switch sensor is used for detecting the real-time rotating speed of the driving wheel and uploading data to the monitoring system; the RFID reader is used for reading on-orbit position data stored by a preset RFID tag on the track, obtaining real-time position data of the conveyor and uploading the real-time position data to the monitoring system; the 360-degree laser radar collects real-time data of the running environment of the conveyor, performs preliminary processing and then uploads the data to the server.
The monitoring system is used for fusing real-time position data fed back by the RFID reader and accumulated mileage calculated according to real-time rotating speed fed back by the proximity switch sensor to position each conveyor with high precision. The system calculates the relative distance between adjacent conveyors and stores real-time speed, voltage, current, load, and two-axis angle status data for each conveyor. At the same time, the system uses 360 ° lidar to collect various environmental information about the circular monorail, including the use of the rail and possible dynamic and static obstructions. By analyzing the data and combining the real-time position of each conveyor, the system calculates and obtains the obstacle avoidance decision of the conveyor according to the obstacle avoidance strategy, and controls the conveyor in real time according to the decision.
Further, the proximity switch sensor is responsible for detecting the number of roller passes of the conveyor drive wheel over a fixed period of time and transmitting this data to the monitoring system. The monitoring system calculates the rotation number of the driving wheel in the time period according to the received data and the known number of rollers contained in the single circle of the driving wheel, so as to further determine the rotation speed of the driving wheel of the conveyor.
Further, the simulated annealing algorithm comprises the following steps:
step 1.1, according to the existing environmental data, integrating the positions and states of all the conveyors and the cargo loading and unloading amount (namely, task amount) requirements of loading and unloading points to set parameters, and setting an initial temperature T_initial, an end temperature T_final and a cooling rate alpha for a simulated annealing algorithm;
step 1.2, generating an initial conveyer task allocation feasible scheme, randomly determining an initial conveyer task sequence, and then performing simulated annealing iteration;
step 1.3, fine tuning the current scheme to generate a new scheme in each iteration process. The fine tuning comprises the steps of changing the execution sequence of two tasks and adjusting parameters of a single task, calculating the cost of a new scheme by utilizing a preset cost function, comparing the cost with the current scheme, selecting whether to adopt the new scheme or not according to the cost difference value and the current temperature, and updating the current temperature by using the cooling rate alpha;
and 1.4, judging whether the preset maximum iteration times are reached, if yes, judging that the iteration is ended, obtaining a conveyor task allocation feasible scheme set, otherwise, turning to the step 1.3, and continuing the iteration.
4. An intelligent hillside orchard transport system according to claim 3, wherein said step 1.2 specifically comprises the steps of:
step 1.21, arranging task set task S in descending order according to task quantity of task points, selecting task point numbers of single task points not smaller than the load of the conveyor, and subtracting corresponding load quantity of the conveyor from the load quantity of the task points to update the load quantity of the conveyor. Repeating the selecting and updating processes until the task quantity of each task point is smaller than the load of a single conveyor;
step 1.22, selecting a task point number combination with the sum of the task amounts of two task points exactly equal to the load, arranging the task point number combination in an ascending order according to the task point numbers, and updating the task amounts of the rest task points; selecting a task point number combination with the sum of the task amounts of the three task points exactly equal to the load, arranging the combination in ascending order according to the task point numbers, and updating the task amounts of the rest task points;
step 1.23, arranging the task set task S in descending order according to the task quantity of the task points, selecting task point combinations with the sum of the task quantities not smaller than the load of a single conveyor according to the sequence, arranging the task point combinations in ascending order according to the task point numbers in the combinations, updating the task quantity of the remaining task points by using a simulated annealing algorithm, and repeatedly sequencing, selecting and updating until the total task quantity is smaller than the load of the single conveyor, wherein the obtained remaining task points are used for distributing a feasible scheme for the tasks of the conveyor.
Further, the ant colony algorithm includes the steps of:
step 2.1, selecting an initial position, wherein each simulated ant randomly selects an initial loading and unloading point as the initial position;
step 2.2, carrying out information exchange and path selection, and collecting real-time task data of states of other nearby conveyors and loading and unloading points by ants through a wireless communication system before moving to the next loading and unloading point;
step 2.3, determining the next destination and running direction of ants by using probability decision rules based on the information collected in the last step and the pheromone concentration, under which the ants evaluate the pheromone concentration and the current traffic condition of each optional destination, then selecting one destination and running direction (clockwise or anticlockwise) according to the determined probability, and then going to the selected loading and unloading points;
step 2.4, at each loading and unloading point, ants not only update their path information, but also release some pheromones at the point to indicate the quality of the path, and transmit the updated path information to nearby conveyors in real time;
step 2.5, after the simulation of the transportation route is completed for all ants, the system carries out global updating of the pheromone and judges whether the preset maximum iteration times are reached; if yes, go to the next step; otherwise, returning to the step 2.1, and starting task simulation of the next round;
and 2.6, outputting an optimal conveyor task execution sequence according to the last transportation route simulation result.
Further, the conveyor is used for loading and unloading at a designated loading and unloading point according to an allocated task, a control system, a current sensor, a voltage sensor, a load sensor, a two-axis angle sensor, a proximity switch sensor, an RFID reader and a 360-degree laser radar are arranged on the conveyor, and the current sensor and the voltage sensor are used for detecting real-time current and real-time voltage of a motor of the conveyor and uploading data to a monitoring system; the load sensor is used for detecting the real-time load capacity of the container and uploading data to the monitoring system; the two-axis angle sensor is used for detecting the left-right swing angle and the real-time upward-downward slope angle of the conveyor and uploading data to the monitoring system; the proximity switch sensor is used for detecting the real-time rotating speed of the driving wheel and uploading data to the monitoring system; the RFID reader is used for reading on-orbit position data stored by a preset RFID tag on the track, obtaining real-time position data of the conveyor and uploading the real-time position data to the monitoring system; the 360-degree laser radar collects real-time data of the running environment of the conveyor, performs preliminary processing and then uploads the data to the server.
After the technical scheme is adopted, compared with the prior art, the invention has the following advantages:
1. the intelligent monitoring, dispatching and transportation task allocation of the multiple mountain transportation machines are integrated, so that the running state of each transportation machine can be monitored and controlled, and all transportation machines on the same track can be subjected to batch unified dispatching control, so that the on-track control efficiency of the multiple transportation machines is remarkably improved.
2. The intelligent task allocation and scheduling algorithm of the conveyor is utilized, the intelligent task allocation and scheduling algorithm is suitable for application scenes of conveying work of a plurality of conveyors on a single track in a clockwise and anticlockwise mode aiming at a plurality of loading and unloading points, and the intelligent task allocation and scheduling algorithm has the characteristic of high scheduling and allocation efficiency.
3. The invention combines two modes to carry out high-precision positioning of the conveyor, and comprises the following steps: (1) the method comprises the steps of counting and recording the rotating speed of a driving wheel of a conveyor in real time, calculating accumulated mileage, and determining the relative position of an annular track where the conveyor is located; (2) and using an RFID reader to read a preset RFID tag on the track, and determining the absolute position of the annular track where the conveyor is located. The combination of the two methods realizes the positioning of the conveyor with low cost and high precision.
4. The invention combines two modes with obstacle avoidance strategies, and aims to realize high-precision and high-efficiency obstacle detection and obstacle avoidance of a conveyor, and the method comprises the following steps:
real-time location sharing policy: (1) and (3) data acquisition: each conveyor is equipped with a high precision positioning system that captures its precise position in the work environment in real time. (2) Data synchronization: the position data is transmitted to the central processing server in real time. (3) Collision prediction: the central server analyzes the relative distance and potential collision path between multiple conveyors in real time through the high-speed calculation module. (4) Obstacle avoidance decision: once a potential collision risk is predicted, the system will quickly develop obstacle avoidance strategies for the relevant conveyors and send control commands to adjust their operating conditions.
Laser radar perception strategy: (1) and (3) data acquisition: the 360-degree laser radar equipped on the conveyor scans the working environment to acquire high-resolution environment information. (2) And (3) data processing: firstly, primarily removing noise by using a filtering algorithm, and ensuring the accuracy of data; then carrying out track characteristic analysis to identify obstacles and abnormal environments on the track; then performing obstacle clustering operation, and identifying and dividing each independent obstacle by using a clustering algorithm based on point cloud data returned by the radar; performing feature extraction to further analyze characteristics of the obstacle, such as size, shape, and direction of movement; and finally, carrying out obstacle avoidance decision, and according to the characteristics and the positions of the obstacles, preparing a proper obstacle avoidance strategy, such as speed adjustment, driving direction change or complete parking waiting.
By combining the two strategies, the invention not only can effectively identify and avoid various obstacles, but also can ensure the safe distance among a plurality of conveyors, thereby remarkably improving the operation safety and the working efficiency of the conveyors.
The invention will now be described in detail with reference to the drawings and examples.
Drawings
FIG. 1 is a circular single track multiple conveyor operating scenario;
FIG. 2 is a schematic side view of a conveyor according to the present invention;
FIG. 3 is a flow chart of a multi-conveyor task scheduling operation;
FIG. 4 is a flow chart of a conveyor "carpool" combination optimization algorithm;
FIG. 5 is a flow chart of a conveyor task allocation algorithm;
FIG. 6 is a flow chart of a process for generating obstacle avoidance decisions.
In the drawings, the list of components represented by the various numbers is as follows:
1. a transport rail; 2. transporting the cargo box; 3. a load sensor; 4. a two-axis angle sensor; 5. a proximity switch sensor; 6. a conveyor drive wheel; 7. a DC motor; 8. 360-degree laser radar; 9. a control cabinet; 10. an RFID tag; 11. an RFID reader; 12. a gearbox; 13. current sensor (inside the control cabinet) 14, voltage sensor (inside the control cabinet)
Detailed Description
The principles and features of the present invention are described below with reference to the drawings, the examples are illustrated for the purpose of illustrating the invention and are not to be construed as limiting the scope of the invention.
1. Multi-conveyor transportation scheduling scene and conditions
The plurality of conveyors run in a single track and can run clockwise or anticlockwise, the numbers of the conveyors and the numbers of loading and unloading points are numbered from small to large clockwise, and the transportation working scene is shown in figure 1. According to mountain transportation demand, every loading and unloading point all has cargo loading and unloading task, can wireless communication between every conveyer and also can carry out information transmission between conveyer and the loading and unloading point, ensures the high efficiency and the security in the transportation.
2. Multi-transporter monitoring content
The multi-conveyor monitoring comprises real-time current, real-time voltage, real-time rotation speed of driving wheels, real-time running speed of the conveyors, real-time two-axis angles of the conveyors, real-time cargo capacity of a container, real-time position of the conveyors and real-time obstacle conditions. The structure of the conveyor is shown in fig. 2, and a control cabinet 9, a current sensor 13, a voltage sensor 14, a proximity switch sensor 5, a two-axis angle sensor 4, a load sensor 3, an RFID reader 11 and a 360-degree laser radar 8 are arranged on the conveyor.
The real-time current and the real-time voltage are uploaded to a monitoring system by real-time data measured by a current and voltage sensor and then are led into a server; the real-time rotating speed of the driving wheel is measured by a proximity switch sensor, the real-time running speed of the conveyor is calculated by the real-time rotating speed, and the data is uploaded to a monitoring system and then is imported into a server; the real-time two-axis angle of the conveyor is measured by a two-axis angle sensor, the two-axis angle data embody the left-right swing angle and the real-time upward-downward slope angle of the conveyor, and the data are uploaded to a monitoring system and then are imported into a server; the real-time load capacity of the container is measured by load sensors at four corners of the container, real-time load data is obtained through calculation of an accumulation module, and the data is uploaded to a monitoring system and then is imported into a server; the real-time position of the conveyor is monitored by a proximity switch sensor, the rotation of the driving wheel is counted, the rotating speed is recorded, the accumulated distance is calculated, the relative position of the annular conveying track (shown in figure 1) is judged, meanwhile, the on-track position data stored by the preset RFID tag 10 on the track is read by the RFID reader 11, the absolute position assignment is carried out on the real-time refreshing relative position of the conveyor, and therefore high-precision positioning is achieved, and the position data is directly uploaded to a server. The real-time obstacle condition is obtained by the 360-degree laser radar 8, laser radar data are uploaded to a server for processing, object detection is carried out, possible obstacles are identified, then obstacle classification (such as people, conveyors, leaves and the like) is carried out, the real-time position of each conveyor is combined according to the type and the size of the obstacle, and obstacle avoidance decisions (such as operation suspension or operation speed and direction adjustment) are comprehensively made.
The real-time position of the conveyor is monitored by a conveyor proximity switch sensor to rotate and count the driving wheel, the rotating speed is recorded in real time, and the accumulated distance is calculated, as shown in fig. 2, the proximity switch sensor is arranged on the driving wheel and used for accurately detecting the passing times of the roller. Based on the fixed distance between the track racks, the running speed and the accumulated running mileage can be calculated, and the on-track relative position can be obtained. Meanwhile, RFID tags arranged on the side edges of the tracks are read by using an RFID reader arranged on the vehicle body, the tags preset track position information, the obtained track position information is updated, the refreshing proximity switch sensor is matched with the driving wheel to obtain on-track relative position information, the position data is uploaded to the server, and the number of the RFID tags is reduced while high-precision positioning is realized, so that the cost is reduced.
The real-time obstacle monitoring condition of the conveyor is obtained by a 360-degree laser radar, and the installation position of the 360-degree laser radar is shown in fig. 2. The data of the radar scan not only covers the location and shape of dynamic obstacles (e.g., people, other conveyers, etc.), but also captures static structures in the environment, such as rails and their supporting steel pipes. These data are first initially processed by a front-end algorithm to identify and exclude fixed rail and support steel pipe structures. The lidar data is then uploaded to a server for further processing. Advanced algorithms on the server first perform object detection, identify potential obstructions, and then classify the obstructions (e.g., people, conveyors, leaves, unexpected obstructions, etc.). The type, size, shape and position data of the obstacle are used to make obstacle avoidance decisions. Decisions include suspending the conveyor, adjusting speed, changing direction, etc. In addition, the real-time location of each conveyor is also taken into account in this decision process, ensuring overall efficiency and safety.
This "front-end algorithm" refers to an algorithm that is executed directly on the hardware of the conveyor (the controller of the lidar or the corresponding embedded system). The purpose of these algorithms is to perform preliminary processing on the raw data to reduce the amount of data, to filter out noise, or to perform some basic feature extraction, etc.
The stage of the front-end algorithm application includes:
1. noise filtering: background noise or anomalous readings are removed from the radar data.
The algorithm is used: kalman filtering
Kalman filtering is a lightweight algorithm suitable for processing continuous data streams in real time, and can effectively remove noise and predict future positions of the conveyor.
2. Identifying a track and a supporting steel pipe: in view of the fact that the rails and support structures are fixed, these structures can be identified by pre-entered shapes or features, or by geometric matching identification algorithms, and excluded from obstacle detection.
The algorithm is used: hough transform
The track has a fixed linear structure and therefore the Hough transform becomes a suitable algorithm for identifying the linear structure. Once the system is properly calibrated at initial installation, the Hough transform can efficiently help the system distinguish between tracks and real obstructions.
3. Primary obstacle recognition: preliminary identification of obvious obstacles may be performed, such as simple detection of large dynamic obstacles, such as other conveyors.
The algorithm is used: shape-based simple thresholding
By setting an appropriate shape threshold, it is possible to distinguish between a background and an obstacle. Segmentation is performed using the size, boundaries, or other geometric characteristics of the object to accurately identify potential obstructions.
The stages of advanced algorithm application on the server include:
1. object detection: faster R-CNN was used. To ensure that the algorithm adapts to the lidar data, model training is performed specifically for the dataset. Specifically, a pre-trained fast R-CNN model is adopted, and the laser radar data with the labels provided in advance is used for training, so that all potential obstacles are more accurately identified.
2. Obstacle classification: once the object is detected, it is classified using the ResNet-50 model. Similar to the object detection phase, a pre-trained ResNet-50 model will be employed and trained using the classified dataset, ensuring that the model can accurately classify people, other conveyors, leaves and other obstacles.
3. Obstacle avoidance decision: this stage will employ a PID controller. The controller may respond to the detected obstacle by halting operation, slowing down or other suitable action, depending on the type, size, shape and location of the obstacle. The PID control parameters are optimized for the dynamic characteristics of the conveyor to ensure the rapidity and accuracy of the response.
3. Multi-conveyor transportation scheduling task flow
The overall flow of the multi-conveyor task scheduling operation is shown in fig. 3, and the specific details are divided into the following steps.
(1) Environment data initialization and transportation task entry
The object is: acquiring current state of system and inputting new transportation task
The conveyor system is started, states such as the positions of all conveyors, the number of on-orbit available conveyors and the like are obtained, a user inputs all loading and unloading numbers and corresponding loading and unloading amounts by utilizing a mobile phone client or a PC (personal computer) end, and task starting execution time is set.
(2) To make 'carpooling' combination
The object is: optimizing a common mission for multiple conveyors
The simulated annealing algorithm is used for carrying out the combination of the carpooling, and the specific flow is shown in figure 4. And integrating the positions, states and cargo demands of loading and unloading points of all the conveyors according to the existing environmental data. And setting parameters, namely setting an initial temperature T_initial, an end temperature T_final and a cooling rate alpha for the simulated annealing algorithm. An initial combination is generated, an initial conveyor mission sequence is randomly determined, then simulated annealing iterations are performed, in each internal loop, the current combination is slightly adjusted to explore a new solution, the new combination cost is calculated using a predetermined cost function and compared with the current combination, and based on the cost difference and the current temperature, whether the new combination is adopted is selected, and the current temperature is updated using the cooling rate alpha. And after the algorithm is finished, obtaining an approximately optimal 'carpooling' scheme.
(3) Task allocation
The object is: assigning specific tasks to each conveyor
The ant colony algorithm is adopted for task allocation, and the specific flow is shown in fig. 5. And (3) carrying out initial position selection, wherein each simulated ant randomly selects an initial loading and unloading point as an initial position, then carrying out information communication and path selection, and collecting the state of other nearby conveyors and real-time task data of the loading and unloading points by the ant through a wireless communication system before moving to the next loading and unloading point. Based on these information and pheromone concentration, ants select the loading and unloading point to which they go next, and decide the clockwise and counterclockwise running directions. And then information updating and transmission are carried out, and ants update the path information of the ants in the system after passing through each loading and unloading point. Meanwhile, the information can be transmitted to a nearby conveyor in real time, so that the real-time performance and the synergy of the whole system are ensured. And (3) cycling and iterating, wherein after all ants complete one simulation, the system performs one global update of the pheromone, and then judges whether the next round of task simulation is required according to the set termination condition. And finally, distributing tasks, and generating an optimal task execution sequence by the system after the termination condition is met.
(4) Implementing tasks
The object is: ensuring accurate execution of tasks by a conveyor
The best task execution sequence obtained by the task allocation stage is transmitted by the server to each conveyor. To ensure that each transporter receives the task sequence correctly, the server waits for confirmation feedback from each transporter. After confirming that there is no error, each conveyor will begin the conveyor work according to this sequence.
(5) Monitoring and feedback
The object is: real-time monitoring task progress, ensuring transportation safety and efficiency
The conveyor monitoring system can monitor the real-time position of each conveyor and the relative position between other conveyors, and comprises the speed, voltage, current and other states of each conveyor, and can collect various environmental information about the annular single track, such as the use condition of the track, possible obstacles and other factors which can influence the task. The user can check the transportation progress and the predicted completion time of each transporter through the mobile phone client or the PC. If the server detects the abnormal behavior or potential failure of a certain conveyor, the server immediately notifies the related conveyor to perform self-checking or stop running, and simultaneously pops up a notification at the user side to inform possible problems and treatment measures. In addition, the server can continuously optimize the scheduling algorithm according to the actual transportation condition and user feedback so as to realize higher transportation efficiency and accuracy in future tasks.
4. Summary
The transport machine monitoring system acquires the data such as the accurate position, cargo carrying capacity, on-orbit barriers and the like of each transport machine, uploads the data to the server end for processing, provides data support for the multi-transport machine cooperative task scheduling work, and further utilizes an algorithm to process real-time tasks issued by users after the scheduling system acquires effective data, so that the on-orbit transport machine can efficiently and safely execute transport tasks.
The foregoing is illustrative of the best mode of carrying out the invention, and is not presented in any detail as is known to those of ordinary skill in the art. The protection scope of the invention is defined by the claims, and any equivalent transformation based on the technical teaching of the invention is also within the protection scope of the invention.
Claims (5)
1. The intelligent hillside orchard transportation system is characterized by comprising a dispatching system, a monitoring system and transportation machines, wherein the dispatching system is used for determining a feasible task distribution scheme of the transportation machines by using a simulated annealing algorithm according to the positions, the number and the task requirements of loading and unloading points on an annular single track, screening the feasible scheme by using an ant colony algorithm to obtain an optimal task distribution scheme, and distributing tasks to each transportation machine according to the optimal task distribution scheme;
the simulated annealing algorithm comprises the following steps:
step 1.1, according to the existing environmental data, integrating the positions and states of all the conveyors and the cargo loading and unloading amount of loading and unloading points, namely, the task amount requirement, setting an initial temperature T_initial, an ending temperature T_final and a cooling rate alpha for a simulated annealing algorithm;
step 1.2, generating an initial conveyer task allocation feasible scheme, randomly determining an initial conveyer task sequence, and then performing simulated annealing iteration;
step 1.3, in each iteration process, fine tuning the current scheme to generate a new scheme; the fine tuning method comprises the steps of changing the execution sequence of two tasks and adjusting parameters of a single task, calculating the cost of a new scheme by utilizing a preset cost function, comparing the cost with the current scheme, selecting whether to adopt the new scheme or not according to the cost difference value and the current temperature, and updating the current temperature by using the cooling rate alpha;
step 1.4, judging whether the preset first maximum iteration times are reached, if yes, judging that the iteration is ended, obtaining a conveyor task allocation feasible scheme set, otherwise, turning to step 1.3, and continuing the iteration; the conveyor is used for loading and unloading at a specified loading and unloading point according to an allocated task, a control system, a current sensor, a voltage sensor, a load sensor, a two-axis angle sensor, a proximity switch sensor, an RFID reader and a 360-degree laser radar are arranged on the conveyor, and the current sensor and the voltage sensor are used for detecting real-time current and real-time voltage of a motor of the conveyor and uploading the real-time current and the real-time voltage to a monitoring system; the load sensor is used for detecting the real-time load capacity of the container and uploading data to the monitoring system; the two-axis angle sensor is used for detecting the left-right swing angle and the real-time upward-downward slope angle of the conveyor and uploading detection data to the monitoring system; the proximity switch sensor is used for detecting the real-time rotating speed of the driving wheel and uploading detection data to the monitoring system; the RFID reader is used for reading on-orbit position data stored by a preset RFID tag on the track, obtaining real-time position data of the conveyor and uploading the real-time position data to the monitoring system; the 360-degree laser radar acquires real-time data of the running environment of the conveyor, performs preliminary processing and then uploads the data to the server;
the monitoring system is used for fusing real-time position data fed back by the RFID reader and accumulated mileage calculated according to real-time rotating speed fed back by the proximity switch sensor to position each conveyor with high precision; calculating the relative distance between adjacent conveyors and storing real-time speed, voltage, current, load and two-axis angle state data of each conveyor; meanwhile, the monitoring system collects various environmental information about the annular single track by using the 360-degree laser radar, including the use condition of the track and possible dynamic and static obstacles; by analyzing the data and combining the real-time position of each conveyor, the monitoring system calculates and obtains the obstacle avoidance decision of the conveyor according to the obstacle avoidance strategy, and controls the conveyor in real time according to the decision.
2. The intelligent hillside orchard transport system of claim 1, wherein the proximity switch sensor is used to detect the number of roller passes of the drive wheel of the transporter over a fixed period of time and send data to the monitoring system; the monitoring system is used for calculating the rotation number of the driving wheel in the time period according to the received data and the known number of rollers contained in a single circle of the driving wheel, so as to determine the rotation speed of the driving wheel of the conveyor.
3. The intelligent hillside orchard transport system according to claim 2, wherein said step 1.2 specifically comprises the steps of:
step 1.21, task set taskS According to the descending order of the task quantity of the task points, selecting the task point number of a single task point, wherein the task quantity of the single task point is not less than the task point number of the conveyor load, and subtracting the corresponding conveyor load quantity from the cargo quantity of the task point to update the cargo quantity; repeating the selecting and updating processes until the task quantity of each task point is smaller than the load of a single conveyor;
step 1.22, selecting a task point number combination with the sum of the task amounts of two task points exactly equal to the load, arranging the task point number combination in an ascending order according to the task point numbers, and updating the task amounts of the rest task points; selecting a task point number combination with the sum of the task amounts of the three task points exactly equal to the load, arranging the combination in ascending order according to the task point numbers, and updating the task amounts of the rest task points;
step 1.23, task is collected from the rest tasksS And (3) arranging the task points in descending order according to the task quantity of the task points, sequentially selecting task point combinations with the sum of the task quantities not smaller than the load of a single conveyor, arranging the task point combinations in ascending order according to the task point numbers in the combinations, updating the task quantity of the remaining task points by using a simulated annealing algorithm, and repeatedly sequencing, selecting and updating until the total task quantity is smaller than the load of the single conveyor, wherein the obtained remaining task points are used for distributing a feasible scheme for the tasks of the conveyor.
4. The intelligent hillside orchard transport system according to claim 1, wherein the ant colony algorithm includes the steps of:
step 2.1, selecting an initial position, wherein each simulated ant randomly selects an initial loading and unloading point as the initial position;
step 2.2, carrying out information exchange and path selection, and collecting real-time task data of states of other nearby conveyors and loading and unloading points by ants through a wireless communication system before moving to the next loading and unloading point;
step 2.3, determining the next destination and running direction of ants by using probability decision rules based on the information collected in the last step and the pheromone concentration, under the rules, the ants evaluate the pheromone concentration and the current traffic condition of each optional destination, then select one destination and running direction according to the determined probability, and then go to the selected loading and unloading points;
step 2.4, at each loading and unloading point, ants not only update their path information, but also release some pheromones at the point to indicate the quality of the path, and transmit the updated path information to nearby conveyors in real time;
step 2.5, after the simulation of the transportation route is completed for all ants, the system carries out global updating of the pheromone and judges whether the preset maximum iteration times are reached; if yes, go to the next step; otherwise, returning to the step 2.1, and starting task simulation of the next round;
and 2.6, outputting an optimal conveyor task execution sequence according to the last transportation route simulation result.
5. The intelligent hillside orchard transport system according to claim 1, wherein the front end of the transporter is equipped with 360 ° lidar for real-time monitoring of the position and shape of dynamic obstacles and static structures; the data acquired by the radar is primarily processed by a front-end algorithm to identify and exclude fixed rails and supporting steel pipe structures, and only dynamic obstacle information is reserved; uploading the processed data to a server for further processing; the algorithm on the server detects the object of the laser radar data, identifies potential obstacles, classifies the obstacles, and obtains the type, size, shape and position data of the obstacles; the data are combined with the real-time position of each conveyor and used for calculating obstacle avoidance decisions of the conveyors, so that the overall high efficiency and safety are ensured.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311275155.1A CN117474422B (en) | 2023-09-28 | 2023-09-28 | Intelligent hillside orchard transportation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311275155.1A CN117474422B (en) | 2023-09-28 | 2023-09-28 | Intelligent hillside orchard transportation system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117474422A CN117474422A (en) | 2024-01-30 |
| CN117474422B true CN117474422B (en) | 2024-04-09 |
Family
ID=89632145
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311275155.1A Active CN117474422B (en) | 2023-09-28 | 2023-09-28 | Intelligent hillside orchard transportation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117474422B (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018176595A1 (en) * | 2017-03-31 | 2018-10-04 | 深圳市靖洲科技有限公司 | Unmanned bicycle path planning method based on ant colony algorithm and polar coordinate transformation |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20220156670A1 (en) * | 2020-11-17 | 2022-05-19 | Deere & Company | Smart orchard harvesting cart with analytics |
| CN113486293B (en) * | 2021-09-08 | 2021-12-03 | 天津港第二集装箱码头有限公司 | Intelligent horizontal transportation system and method for full-automatic side loading and unloading container wharf |
-
2023
- 2023-09-28 CN CN202311275155.1A patent/CN117474422B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018176595A1 (en) * | 2017-03-31 | 2018-10-04 | 深圳市靖洲科技有限公司 | Unmanned bicycle path planning method based on ant colony algorithm and polar coordinate transformation |
Non-Patent Citations (2)
| Title |
|---|
| Development and tests of sliding contact line-powered track transporter;Yihua Jiang 等;《Int J Agric & Biol Eng 》;20230731;第16卷(第4期);第68–75页 * |
| 丘陵山地环形单轨运输系统静态调度优化方法;杨方 等;《农业工程学报》;20230228;第39卷(第4期);第37-46页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117474422A (en) | 2024-01-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112149555B (en) | Global vision-based multi-warehouse AGV tracking method | |
| Halawa et al. | Introduction of a real time location system to enhance the warehouse safety and operational efficiency | |
| Sabattini et al. | The pan-robots project: Advanced automated guided vehicle systems for industrial logistics | |
| US10083418B2 (en) | Distributed autonomous robot systems and mehtods | |
| CN110371112A (en) | A kind of intelligent barrier avoiding system and method for automatic driving vehicle | |
| CN111768629B (en) | Vehicle scheduling method, device and system | |
| CN107610282A (en) | A kind of bus passenger flow statistical system | |
| CN111738323A (en) | Mixed enhanced intelligent track prediction method and device based on gray Markov model | |
| CN114030805A (en) | Warehousing system, shuttle vehicle for warehousing system and navigation method of shuttle vehicle | |
| CN115759492B (en) | Warehouse management method and system based on convolutional neural network algorithm | |
| US20230111354A1 (en) | Method and system for determining a mover model for motion forecasting in autonomous vehicle control | |
| KR102119161B1 (en) | Indoor position recognition system of transpotation robot | |
| US20230147624A1 (en) | Autonomous mobile robots for coverage path planning | |
| CN117474422B (en) | Intelligent hillside orchard transportation system | |
| CN117234214A (en) | Automatic shuttle for stacking industrial goods | |
| US11183052B2 (en) | Enhanced vehicle operation | |
| CN116382292A (en) | Multi-machine collaborative scheduling method, equipment and medium on cloud of moving robot | |
| CN115373395A (en) | Four-way vehicle scheduling method, system and equipment | |
| CN115188192A (en) | Automatic control parking system and method based on travel prediction | |
| CN115293606A (en) | Unmanned delivery vehicle system | |
| CN114510037A (en) | Intelligent road condition analysis method and system based on unmanned driving | |
| CN117735414B (en) | Speed control system and method for wafer carrying system | |
| Xiang et al. | A study of the city bus passenger flow intelligent statistical algorithm based on motion vector tracking | |
| US20230385684A1 (en) | Machine learning for management of positioning techniques and radio frequency usage | |
| CN117893115A (en) | Efficient intelligent warehouse transportation system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |