Intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion and control method

Abstract

The invention relates to an intelligent omni-directional logistics vehicle based on multi-source heterogeneous information fusion and a control method. The logistics vehicle comprises a logistics vehicle body, an integrated sensor box module, a visual module, a control module and a driving module, wherein the integrated sensor box module, the visual module, the control module and the driving module are arranged on the logistics vehicle body. The integrated sensor box module, the vision module, the control module and the driving module are all in communication connection with the control module; the control module obtains image information collected by the monocular camera and the depth camera for mapping, plans a driving route and finds an optimal solution of a cargo circulation path; the control module obtains data of the left sensor integration box and the right sensor integration box for integration analysis, judges the conditions of obstacles on two wings and in front of the side, and assists camera information in obstacle avoidance. And the control module acquires data of the inertial measurement unit, and performs visual SLAM (Simultaneous Localization and Mapping) processing by fusing image information. Compared with the prior art, high-precision mapping is carried out, goods are dispatched efficiently, and the obstacle avoidance and path planning capability of the logistics vehicle is improved.

Description

Technical Field

[0001] The present invention relates to an intelligent logistics vehicle, and in particular to an intelligent omnidirectional logistics vehicle and a control method based on multi-source heterogeneous information fusion. Background Art

[0002] In today's era, with the rapid development of the logistics and warehousing industry, the drawbacks of traditional manual warehousing are gradually exposed. In the traditional model, relying on manpower for cargo turnover is the main mode of operation of logistics and warehousing, including but not limited to cargo loading and unloading, sorting and dispatching. The traditional business model has the problems of high labor costs and low efficiency, and is limited by the characteristics of human resources. The fault tolerance of this model is far from that of intelligent logistics and warehousing. High-tech is an important pillar of the modernization of logistics and warehousing, and is an indispensable part of realizing the transformation from digital warehousing to smart warehousing.

[0003] With the development of online shopping platforms, people's demand for online shopping is getting stronger and stronger, and with it comes the increasing pressure of warehousing and logistics in various regions. In order to ensure the delivery speed and accuracy of logistics warehouses, the demand for real-time and accurate cargo flow in warehousing and logistics is becoming stronger and stronger. The intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion integrates the information of different sensors, which to a certain extent avoids the perception limitations of a single sensor and the instability of information transmission. The intelligent logistics vehicle with multi-source heterogeneous information fusion realizes a more comprehensive recognition and perception of the environment, helps the transformation of warehousing and logistics to modernization and intelligence, and provides a flexible solution for the optimization of traditional manual logistics warehousing.

[0004] Mecanum wheels are all-round wheels suitable for flat roads. Through the independent control of the four wheels by the main control chip, they can achieve omnidirectional movement, thus solving the problem of insufficient driving lanes for smart logistics vehicles in the warehousing and logistics process due to the narrow space. They are very suitable for logistics scheduling in warehousing environments. They can greatly improve the efficiency of warehousing space logistics transportation, the utilization rate of logistics warehousing space, and reduce labor costs.

[0005] The core technology of intelligent omnidirectional logistics vehicles is simultaneous localization and mapping (SLAM). In the existing technology, laser SLAM is often used for mapping logistics vehicles, while the mapping accuracy of single light radar is not high and it is easily affected by the environment, and the fault tolerance performance is poor. Summary of the invention

[0006] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide an intelligent omnidirectional logistics vehicle and control method based on multi-source heterogeneous information fusion, integrate visual SLAM and inertial measurement unit into the omnidirectional logistics vehicle, and at the same time, IMU adopts MPU9250 chip solution to perform high-precision mapping, efficiently dispatch goods, and improve the obstacle avoidance and path planning capabilities of the logistics vehicle.

[0007] Depth cameras and monocular cameras are used to build environmental scene maps based on visual SLAM. At the same time, a multi-sensor integrated box is equipped to assist omnidirectional logistics vehicles in predictable track changes and obstacle avoidance. This can improve the vehicle structure, improve the positioning and mapping accuracy of logistics vehicles, and enrich the diversity of intelligent logistics vehicles to a certain extent.

[0008] The present invention provides an intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion, comprising: a logistics vehicle body and an integrated sensor box module, a vision module, a control module, and a drive module arranged on the logistics vehicle body; the integrated sensor box module, the vision module, the control module, and the drive module are all communicatively connected with the control module;

[0009] The visual module includes: a monocular camera, a depth camera, and an inertial measurement unit, which are used to simultaneously map and locate the surrounding environment; the integrated sensor box module includes: a left sensor integrated box and a right sensor integrated box, and the integrated sensor box module is loaded with a millimeter wave radar and an ultrasonic sensor, which respectively detect different types of environmental data for obstacle identification, and assist the control module in path planning and real-time obstacle avoidance to ensure vehicle driving safety;

[0010] The driving module includes: a left front Mecanum wheel, a left rear Mecanum wheel, a right front Mecanum wheel and a right rear Mecanum wheel, which are used to realize omnidirectional driving functions such as turning around on the spot and moving sideways of the intelligent logistics vehicle;

[0011] The control module obtains the image information collected by the monocular camera and the depth camera for mapping, and plans the driving route at the same time to find the optimal solution for the cargo flow path; the control module obtains the data of the left sensor integrated box and the right sensor integrated box for integrated analysis, determines the obstacle conditions on the two wings and the front side, and assists the camera information for obstacle avoidance; the control module obtains the inertial measurement unit data and fuses the image information for visual SLAM processing.

[0012] Furthermore, the top layer of the logistics vehicle is provided with an anti-skid damping layer made of water-based environmentally friendly material, in which a short-range high-frequency radio communication module and a pressure-sensitive sensor are embedded for NFC information recognition and pressure sensing of the goods placed on the roof.

[0013] Furthermore, the logistics vehicle also includes four Mecanum wheel drive motors TB, namely a left front Mecanum wheel drive motor, a right front Mecanum wheel drive motor, a right rear Mecanum wheel drive motor, and a left rear Mecanum wheel drive motor.

[0014] Furthermore, a battery compartment is provided on the chassis of the logistics vehicle. The battery compartment is located at the rear of the logistics vehicle and is open at one end to facilitate charging or replacing batteries.

[0015] Furthermore, it also includes: an early warning module, when the battery of the logistics vehicle is too low, the early warning module will control the passive buzzer to alarm.

[0016] Furthermore, it also includes: a left front positioning indicator light, a right front positioning indicator light, a right rear positioning indicator light, and a left rear positioning indicator light. When the battery of the logistics vehicle is too low, the left front positioning indicator light, the right front positioning indicator light, the right rear positioning indicator light, and the left rear positioning indicator light turn red.

[0017] Furthermore, it also includes: a communication module. When the battery of the logistics vehicle is too low, the communication module reminds the center administrator to replace the battery. The replacement process takes a short time. After the replacement, the logistics vehicle can immediately carry out logistics tasks, thereby realizing all-weather logistics scheduling.

[0018] Furthermore, when the cargo information cannot be detected normally through the short-range high-frequency radio communication module and the pressure-sensitive sensor or the pressure sensing is abnormal, the early warning module is triggered, and the left front positioning indicator light, the right front positioning indicator light, the right rear positioning indicator light, and the left rear positioning indicator light up red, and the passive buzzer sounds an alarm.

[0019] The present invention also provides a control method based on multi-source heterogeneous information fusion for an intelligent omnidirectional logistics vehicle, characterized in that the logistics vehicle obtains cargo dispatch information according to a short-range high-frequency radio communication module, simultaneously builds a map and locates with the assistance of a visual module, and an integrated sensor box module is provided with a millimeter-wave radar and an ultrasonic sensor to provide environmental information on both wings. After the collected information is transmitted to the control module, the control module fuses the multi-source heterogeneous data to make a comprehensive decision;

[0020] The specific control process includes:

[0021] S1: Extract data information from the front camera monocular camera and depth camera, and integrate the inertial measurement unit for simultaneous positioning and mapping;

[0022] S2: The control module determines the location of the target point based on the cargo flow information obtained by the short-range high-frequency radio communication module and the constructed map, and preliminarily performs local optimal path planning; the control module controls the four Mecanum wheels to move along the planned path;

[0023] S3: During the driving process of the logistics vehicle, the control module can freely dispatch the integrated sensor box module as needed, enable the millimeter-wave radar to analyze the relative position of the flanking vehicles, and make reasonable avoidance of obstacles or abnormally moving vehicles;

[0024] S4: When the vehicle moves longitudinally between narrow shelves, the ultrasonic sensor-assisted control system is activated to analyze the relative position of obstacles;

[0025] S5: The sensor box module communicates with the control module. The sensor box module can change the path planned by the system, lock the drive module, and force the logistics vehicle to change track or standby;

[0026] S6: The control module controls the Mecanum wheel motor to directly drive the logistics vehicle. By controlling the speed and steering parameters of the four independent motors, it can achieve omnidirectional movement including turning around on the spot, longitudinal and diagonal movement.

[0027] Furthermore, the logistics vehicle obtains information through a short-range high-frequency radio communication module, and simultaneously locates and maps the storage space based on visual SLAM to transport the goods to a designated area;

[0028] The following steps are involved:

[0029] A1: Turn on the power supply, and each component starts the module self-test. If the self-test is normal, go to step A3, otherwise go to step A2;

[0030] A2: Self-check is abnormal, the smart logistics vehicle fails to start, and the early warning module triggers the alarm. The passive buzzer sounds an alarm, and the left front positioning indicator, right front positioning indicator, right rear positioning indicator, and left rear positioning indicator are set to red; at the same time, the communication module uploads the fault information to the center administrator, waiting for the fault to be eliminated;

[0031] A3: The self-check is normal, the smart logistics vehicle starts successfully, and each module starts working; the monocular camera and the depth camera start to collect environmental information, and the left sensor integration box and the right sensor integration box immediately feedback the obstacle information on both wings to the control module, and drive to the cargo transfer point. At this time, the top NFC module is in a waiting state. When the pressure-sensitive sensor senses the placement of the cargo, execute step A4;

[0032] A4: The pressure-sensitive sensor detects the mass of the object. If the object is overloaded, execute step A5. Otherwise, execute step A6.

[0033] A5: If the weight of the cargo exceeds the carrying capacity of the logistics vehicle or the NFC data recognition is abnormal, the positioning indicator light is set to red and the passive buzzer sounds an alarm. The alarm is lifted when the overloaded cargo is removed or the NFC data of the non-overloaded cargo is successfully extracted;

[0034] A6: If the weight of the goods is within the acceptable range of the logistics vehicle and the NFC data can be recognized normally, the positioning indicator light is set to green; after reading the NFC data, the route is planned, and the logistics vehicle starts after the passive buzzer sounds a prompt, and the loaded goods are transferred to the target location;

[0035] A7: Integrate multi-source heterogeneous information for path planning and intelligent obstacle avoidance during driving;

[0036] A8: After the dispatch is completed, the communication module will feedback the logistics dispatch information and start the next round of cargo dispatch after returning to the cargo transfer point.

[0037] Compared with the prior art, the present invention has the following advantages:

[0038] (1) Visual SLAM and inertial measurement units are integrated into omnidirectional logistics vehicles. Visual SLAM is used to identify obstacles in the environment. At the same time, the IMU uses the MPU9250 chip solution to provide vehicle dynamic posture information to optimize path planning and obstacle avoidance. In addition, even in narrow shelf aisles where visual information is partially blocked, IMU data can help the vehicle maintain the correct driving posture.

[0039] (2) The Mecanum wheel is used in the transfer of goods in logistics warehousing. Its omnidirectional mobility greatly improves the flexibility of logistics vehicle transportation in narrow spaces, and plays an important role in optimizing warehouse space utilization, labor costs in logistics warehousing, and efficient scheduling of goods.

[0040] (3) An anti-skid damping layer is designed on the top to provide anti-skid protection and slow down the inertia of goods, reducing the risk of goods rolling. In addition, a short-range high-frequency radio communication module is embedded in the top to ensure that the logistics vehicle can accurately obtain the dispatch information of the goods to be circulated in a very short time, thereby improving the efficiency of goods circulation. At the same time, a pressure-sensitive sensor is provided to enhance the data dimension of the high-speed logistics vehicle and assist the logistics vehicle in the storage space to carry out dispatch work.

[0041] (4) Multifunctional use of short-range high-frequency radio communication module. In addition to identifying cargo-related data, the short-range high-frequency radio communication module also has an NFC data storage function, which can store the logistics vehicle's own relevant data (such as parts replacement status, vehicle maintenance time, etc.) in the NFC module for the center administrator to read when needed.

[0042] (5) An obstacle avoidance solution that combines vision and sensor integration boxes is adopted. The millimeter-wave radar and ultrasonic sensor are integrated into the obstacle avoidance solution to monitor the information of obstacles on the logistics vehicle's route in real time. Combined with visual image data, the logistics vehicle can more clearly judge the current path status and perform real-time route planning, which improves the obstacle avoidance and path planning capabilities of the logistics vehicle to a certain extent.

[0043] (6) Sensor integration box, which integrates millimeter-wave radar and ultrasonic sensor. Millimeter-wave radar can provide information such as the speed and angle of the target object, and ultrasonic sensor can provide distance information of the object. By effectively fusing the millimeter-wave radar and ultrasonic sensor data through data fusion algorithms such as Kalman filtering, more accurate road condition analysis and obstacle avoidance can be obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a structural diagram of an intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion;

[0045] Figure 2 A top view of an intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion;

[0046] Figure 3 It is a structural schematic diagram of the integrated sensor box module;

[0047] Figure 4 This is a structural diagram of the logistics vehicle chassis.

[0048] Figure numerals: 1-logistics vehicle body, 2-left front positioning indicator light, 3-left front Mecanum wheel, 4-monocular camera, 5-left front Mecanum wheel drive motor, 6-depth camera, 7-right front positioning indicator light, 8-right front Mecanum wheel drive motor, 9-right front Mecanum wheel, 10-control module, 11-inertial measurement unit, 12-right sensor integration box, 13-battery compartment, 14-right rear Mecanum wheel, 15-right rear Mecanum wheel drive motor, 16-right rear positioning indicator light, 17-passive buzzer, 18-left rear Mecanum wheel drive motor, 19-left rear positioning indicator light, 20-left rear Mecanum wheel, 21-anti-slip damping layer, 22-left sensor integration box, 23-short-range high-frequency radio communication module, 24-pressure sensitive sensor, 25-early warning module, 26-communication module, 27-chassis, 28-millimeter wave radar, 29-ultrasonic sensor. DETAILED DESCRIPTION

[0049] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. Component models, material names, connection structures, control methods, algorithms and other features not clearly described in this technical solution are all considered to be common technical features disclosed in the prior art.

[0050] Example 1

[0051] This embodiment provides an intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion. Figure 1 , 2 , 3, 4, comprising: a logistics vehicle body 1 and an integrated sensor box module, a visual module, a control module 10, and a drive module arranged on the logistics vehicle body; the integrated sensor box module, the visual module, the control module 10, and the drive module are all connected to the control module 10 for communication;

[0052] The visual module includes: a monocular camera 4 and a depth camera 6, and an inertial measurement unit 11, which are used to simultaneously map and locate the surrounding environment; the integrated sensor box module includes: a left sensor integrated box 22 and a right sensor integrated box 12, and the integrated sensor box module is loaded with a millimeter wave radar 28 and an ultrasonic sensor 29, which respectively detect different types of environmental data for obstacle identification, and assist the control module 10 in path planning and real-time obstacle avoidance to ensure vehicle driving safety;

[0053] The driving module includes: a left front Mecanum wheel 3, a left rear Mecanum wheel 20, a right front Mecanum wheel 9 and a right rear Mecanum wheel 14, which are used to realize omnidirectional driving functions such as turning around on the spot and moving sideways of the intelligent logistics vehicle;

[0054] The control module 10 obtains the image information collected by the monocular camera 4 and the depth camera 6 for mapping, and plans the driving route at the same time to find the optimal solution for the cargo flow path; the control module 10 obtains the data of the left sensor integration box 22 and the right sensor integration box 12 for integrated analysis, determines the obstacle conditions on the two wings and the side front, and assists the camera information to avoid obstacles; the control module 10 obtains the data of the inertial measurement unit 11, and fuses the image information for visual SLAM processing.

[0055] In a specific embodiment, the top layer of the logistics vehicle is provided with an anti-skid damping layer 21 made of water-based environmentally friendly material, in which a short-range high-frequency radio communication module 23 and a pressure-sensitive sensor 24 are embedded, which are used to perform NFC information recognition and pressure sensing of the goods placed on the roof.

[0056] In a specific embodiment, the logistics vehicle further includes four Mecanum wheel drive motors TB6612, namely, a left front Mecanum wheel drive motor 5, a right front Mecanum wheel drive motor 8, a right rear Mecanum wheel drive motor 15, and a left rear Mecanum wheel drive motor 18.

[0057] In a specific embodiment, a battery compartment 13 is provided on the chassis 27 of the logistics vehicle. The battery compartment 13 is located at the rear of the logistics vehicle and is open at one end to facilitate charging or replacing batteries.

[0058] In a specific implementation, it also includes: an early warning module 25. When the battery power of the logistics vehicle is too low, the early warning module 25 will control the passive buzzer 17 to sound an alarm.

[0059] In a specific implementation, it also includes: a left front positioning indicator light 2, a right front positioning indicator light 7, a right rear positioning indicator light 16, and a left rear positioning indicator light 19. When the battery power of the logistics vehicle is too low, the left front positioning indicator light 2, the right front positioning indicator light 7, the right rear positioning indicator light 16, and the left rear positioning indicator light 19 will light up red.

[0060] In a specific implementation, it also includes: a communication module 26. When the battery of the logistics vehicle is too low, the communication module 26 reminds the center administrator to replace the battery. The replacement process takes a short time. After the replacement, the logistics vehicle can immediately carry out logistics tasks, thereby realizing all-weather logistics scheduling.

[0061] In a specific embodiment, when the cargo information cannot be normally detected by the short-range high-frequency radio communication module 23 and the pressure-sensitive sensor 24 or the pressure sensing is abnormal, the early warning module 25 is triggered, the left front positioning indicator light 2, the right front positioning indicator light 7, the right rear positioning indicator light 16, and the left rear positioning indicator light 19 light up in red, and the passive buzzer 17 emits an alarm sound.

[0062] The present invention also provides a control method based on multi-source heterogeneous information fusion for an intelligent omnidirectional logistics vehicle, characterized in that the logistics vehicle obtains cargo dispatch information according to a short-range high-frequency radio communication module 23, and simultaneously builds a map and locates with the assistance of a visual module. A millimeter-wave radar 28 and an ultrasonic sensor 29 are provided in the integrated sensor box module to provide environmental information on both wings. After the collected information is transmitted to the control module 10, the control module 10 fuses the multi-source heterogeneous data to make a comprehensive decision.

[0063] The specific control process includes:

[0064] S1: extracting data information from the front camera monocular camera 4 and the depth camera 6, and integrating the inertial measurement unit 11 to perform simultaneous positioning and mapping;

[0065] S2: The control module 10 determines the location of the target point based on the cargo flow information obtained by the short-range high-frequency radio communication module 23 and the constructed map and preliminarily performs local optimal path planning; the control module 10 controls the four Mecanum wheels to move along the planned path;

[0066] S3: During the driving process of the logistics vehicle, the control module 10 can freely dispatch the integrated sensor box module as needed, activate the millimeter wave radar 28 to analyze the relative position of the flanking vehicles, and make reasonable avoidance of obstacles or abnormally moving vehicles;

[0067] S4: When the vehicle moves longitudinally between narrow shelves, the ultrasonic sensor 29 is activated to assist the control system in analyzing the relative position of obstacles;

[0068] S5: The sensor box module communicates with the control module 10, and the sensor box module can change the path planned by the system, lock the drive module, and force the logistics vehicle to change track or standby;

[0069] S6: The control module 10 controls the Mecanum wheel motor to directly drive the logistics vehicle. By controlling the speed and steering parameters of the four independent motors, omnidirectional movement including turning around on the spot, longitudinal movement, and diagonal movement can be achieved.

[0070] The logistics vehicle obtains information according to the short-range high-frequency radio communication module 23, and simultaneously locates and maps the storage space based on visual SLAM to transport the goods to the designated area;

[0071] The following steps are involved:

[0072] A1: Turn on the power supply, and each component starts the module self-test. If the self-test is normal, go to step A3, otherwise go to step A2;

[0073] A2: Self-check is abnormal, the smart logistics vehicle fails to start, and the early warning module 25 triggers an alarm. The passive buzzer 17 sounds an alarm, and the left front positioning indicator light 2, the right front positioning indicator light 7, the right rear positioning indicator light 16, and the left rear positioning indicator light 19 are set to red; at the same time, the communication module 26 uploads the fault information to the center administrator and waits for the fault to be eliminated;

[0074] A3: The self-check is normal, the smart logistics vehicle starts successfully, and each module starts working; the monocular camera 4 and the depth camera 6 start to collect environmental information, and the left sensor integration box 22 and the right sensor integration box 12 immediately feedback the obstacle information on both wings to the control module 10, and drive to the cargo transfer point. At this time, the top NFC module is in a waiting state. When the pressure-sensitive sensor 24 senses the placement of the cargo, execute step A4;

[0075] A4: The pressure-sensitive sensor 24 detects the mass of the object. If the object is overloaded, step A5 is executed. Otherwise, step A6 is executed.

[0076] A5: If the weight of the cargo exceeds the carrying capacity of the logistics vehicle or the NFC data recognition is abnormal, the positioning indicator light is set to red and the passive buzzer 17 sounds an alarm. The alarm is lifted when the overloaded cargo is removed or the NFC data of the non-overloaded cargo is successfully extracted;

[0077] A6: If the weight of the goods is within the range that the logistics vehicle can bear and the NFC data can be recognized normally, the positioning indicator light is set to green; after reading the NFC data, the route is planned, and the logistics vehicle starts after the passive buzzer 17 emits a prompt sound, and the loaded goods are transferred to the target location;

[0078] A7: Integrate multi-source heterogeneous information for path planning and intelligent obstacle avoidance during driving;

[0079] A8: After the dispatch is completed, the communication module 26 feeds back the logistics dispatch information and starts the next round of cargo dispatch after returning to the cargo transfer point.

[0080] In a specific implementation method, machine learning and deep learning can also be used. After the logistics vehicle completes the construction of the regional map, the map data will be shared to a remote computer to grasp the map details and the location information of the logistics vehicle in real time. After deploying multiple logistics vehicles, the large map can be updated synchronously, and the path planning information data can be upgraded to better realize cargo scheduling in large areas.

[0081] In a specific implementation, this intelligent omnidirectional logistics vehicle can also be loaded with other positioning systems, such as a UWB positioning system, which can be combined with an IMU to assist the logistics vehicle in intelligently positioning the vehicle itself and the destination. In the case of vehicle interconnection, the location information of multiple vehicles can also be integrated for joint route planning.

[0082] In a specific implementation, the logistics vehicle can also capture real-time regional information through a high-definition camera, use the YOLO (You Only Look Once) target detection algorithm combined with the regional detection model to achieve environmental anomaly warning, and display the detection results on the computer screen of the center manager. Combined with the synchronized status condition data, the center manager's computer screen will display the data and abnormal points in the area captured by the logistics vehicle.

[0083] Autonomous navigation and obstacle avoidance: Obstacle data of the environment is obtained in real time through the sensor integration boxes on the left and right sides of the logistics vehicle. The environmental point cloud data of the depth camera 6 and the monocular camera 4 are integrated, and the robot's own motion data is obtained in combination with the inertial measurement unit 11. The SLAM algorithm based on the extended Kalman filter creates an incremental environmental map, and uses the map to achieve self-positioning and navigation. The improved A* algorithm based on global planning and local optimization realizes the robot's incremental composition and optimizes the path, completes autonomous obstacle avoidance and path planning, and achieves freedom of movement in complex environments such as narrow warehouses and shopping malls.

[0084] When the intelligent omnidirectional logistics vehicle fails or detects an abnormal environment, the logistics vehicle will upload the fault point to the central administrator for processing. If the vehicle itself fails, the fault module information will be uploaded synchronously. If the environment is abnormal, the abnormal point judgment basis and the fault point location information will be uploaded. At this time, the abnormal point mark signal will appear on the grid map of the central administrator's computer. The central administrator can quickly find the abnormal target point based on the abnormal point location information and the guidance of the grid map to eliminate and correct the abnormal cause.

[0085] When the intelligent omnidirectional logistics vehicle encounters low battery during work, the logistics vehicle uploads the vehicle status and low battery warning information to the central administrator through the communication module. The administrator will dispatch the nearest logistics vehicle to take over the work. At the same time, the low-battery logistics vehicle will automatically go to the central management office, and the central management personnel will perform the battery replacement operation. After completing the charging and battery replacement work, the logistics vehicle will leave the station and continue to complete the cargo dispatch work.

[0086] Components not described in detail in this embodiment are all existing components that can be purchased through public channels.

[0087] The above description of the embodiments is to facilitate the understanding and use of the invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to these embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above embodiments, and improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope of the present invention should be within the scope of protection of the present invention.

Claims

1. An intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion, characterized in that: include: A logistics vehicle body (1) and an integrated sensor box module, a visual module, a control module (10), and a drive module arranged on the logistics vehicle body; the integrated sensor box module, the visual module, the control module (10), and the drive module are all communicatively connected to the control module (10); The visual module comprises: a monocular camera (4), a depth camera (6), and an inertial measurement unit (11), which are used for simultaneously mapping and positioning the surrounding environment; the integrated sensor box module comprises: a left sensor integrated box (22) and a right sensor integrated box (12), wherein the integrated sensor box module is loaded with a millimeter wave radar (28) and an ultrasonic sensor (29), which respectively detect different types of environmental data for obstacle identification, and assist the control module (10) in path planning and real-time obstacle avoidance, thereby ensuring the driving safety of the vehicle; The driving module comprises: a left front Mecanum wheel (3), a left rear Mecanum wheel (20), a right front Mecanum wheel (9) and a right rear Mecanum wheel (14), and is used to realize the omnidirectional driving function of the intelligent logistics vehicle; The control module (10) acquires image information collected by the monocular camera (4) and the depth camera (6) to build a map, and plans a driving route to find the optimal solution for the cargo circulation path; the control module (10) acquires data from the left sensor integration box (22) and the right sensor integration box (12) to integrate and analyze, judge the obstacle conditions on the two wings and the side front, and use the auxiliary camera information to avoid obstacles; the control module (10) acquires data from the inertial measurement unit (11), fuses the image information and performs visual SLAM processing.

2. The intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion according to claim 1 is characterized in that: The top layer of the logistics vehicle is provided with an anti-skid damping layer (21) made of water-based environmentally friendly material, and a short-range high-frequency radio communication module (23) and a pressure-sensitive sensor (24) are embedded in the anti-skid damping layer (21) for performing NFC information recognition and pressure sensing on the goods placed on the roof.

3. The intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion according to claim 1 is characterized in that: The logistics vehicle also includes four Mecanum wheel drive motors TB6612, namely a left front Mecanum wheel drive motor (5), a right front Mecanum wheel drive motor (8), a right rear Mecanum wheel drive motor (15), and a left rear Mecanum wheel drive motor (18).

4. The intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion according to claim 1 is characterized in that: A battery compartment (13) is provided on the chassis (27) of the logistics vehicle. The battery compartment (13) is located at the rear of the logistics vehicle and has an opening at one end, so as to facilitate charging or replacing batteries.

5. The intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion according to claim 1 is characterized in that: Also includes: The early warning module (25) controls the passive buzzer (17) to sound an alarm when the battery level of the logistics vehicle is too low.

6. The intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion according to claim 1 is characterized in that: Also includes: The left front positioning indicator light (2), the right front positioning indicator light (7), the right rear positioning indicator light (16), and the left rear positioning indicator light (19) are lit red when the battery power of the logistics vehicle is too low.

7. The intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion according to claim 1 is characterized in that: Also includes: Communication module (26), when the battery of the logistics vehicle is too low, the communication module (26) reminds the center administrator to replace the battery, and the replacement process takes a short time.

8. The intelligent omnidirectional logistics vehicle based on multi-source heterogeneous information fusion according to claim 1 is characterized in that: When the cargo information cannot be detected normally or the pressure sensing is abnormal, the early warning module (25) is triggered, the left front positioning indicator light (2), the right front positioning indicator light (7), the right rear positioning indicator light (16), and the left rear positioning indicator light (19) are lit red, and the passive buzzer (17) emits an alarm sound.

9. A control method based on multi-source heterogeneous information fusion for an intelligent omnidirectional logistics vehicle as claimed in any one of claims 1 to 8, characterized in that: The logistics vehicle obtains cargo dispatch information based on the short-range high-frequency radio communication module (23), and simultaneously builds a map and locates with the assistance of the visual module. The integrated sensor box module is provided with a millimeter-wave radar (28) and an ultrasonic sensor (29) to provide environmental information on both wings. After the collected information is transmitted to the control module (10), the control module (10) integrates multi-source heterogeneous data and makes a comprehensive decision; The specific control process includes: S1: extracting data information from the front camera monocular camera (4) and the depth camera (6), and integrating the inertial measurement unit (11) to perform simultaneous positioning and mapping; S2: The control module (10) determines the location of the target point based on the cargo flow information obtained by the short-range high-frequency radio communication module (23) and integrates the constructed map and preliminarily performs local optimal path planning; the control module (10) controls the four Mecanum wheels to move along the planned path; S3: During the driving process of the logistics vehicle, the control module (10) can freely dispatch the integrated sensor box module as needed, activate the millimeter wave radar (28) to analyze the relative position of the flanking vehicles, and make reasonable avoidance of obstacles or abnormally moving vehicles; S4: When the vehicle moves longitudinally between narrow shelves, the ultrasonic sensor (29) is activated to assist the control system in analyzing the relative position of obstacles; S5: The sensor box module communicates with the control module (10), and the sensor box module can change the path planned by the system, lock the drive module, and force the logistics vehicle to change track or standby; S6: The control module (10) controls the Mecanum wheel motor to directly drive the logistics vehicle, and realizes omnidirectional movement including turning around on the spot, longitudinal movement, and diagonal movement by controlling the speed and steering parameters of the four independent motors.

10. The control method based on multi-source heterogeneous information fusion of the intelligent omnidirectional logistics vehicle according to claim 9 is characterized in that: The logistics vehicle obtains information according to the short-range high-frequency radio communication module (23), and simultaneously locates and maps the storage space based on visual SLAM to transport the goods to the designated area; The following steps are involved: A1: Turn on the power supply, and each component starts the module self-test. If the self-test is normal, go to step A3, otherwise go to step A2; A2: Self-check is abnormal, the smart logistics vehicle fails to start, and the early warning module (25) triggers an alarm; the passive buzzer (17) sounds an alarm, and the left front positioning indicator light (2), the right front positioning indicator light (7), the right rear positioning indicator light (16), and the left rear positioning indicator light (19) are set to red; at the same time, the communication module (26) uploads the fault information to the center administrator, waiting for the fault to be eliminated; A3: The self-check is normal, the intelligent logistics vehicle starts successfully, and each module starts working; the monocular camera (4) and the depth camera (6) start to collect environmental information, the left sensor integration box (22) and the right sensor integration box (12) immediately feedback the obstacle information on both wings to the control module (10), and the vehicle drives to the cargo transfer point. At this time, the top NFC module is in a waiting state. When the pressure-sensitive sensor (24) senses that the cargo is placed, step A4 is executed; A4: The pressure-sensitive sensor (24) detects the mass of the object. If the object is overloaded, step A5 is executed. Otherwise, step A6 is executed. A5: If the weight of the cargo exceeds the carrying capacity of the logistics vehicle or the NFC data recognition is abnormal, the positioning indicator light is set to red and the passive buzzer (17) sounds an alarm. The alarm is lifted when the overloaded cargo is removed or the NFC data of the non-overloaded cargo is successfully extracted; A6: If the weight of the cargo is within the acceptable range of the logistics vehicle and the NFC data can be recognized normally, the positioning indicator light is set to green; after reading the NFC data, the route is planned, and after the passive buzzer (17) sounds a prompt, the logistics vehicle starts and transfers the loaded cargo to the target location; A7: Integrate multi-source heterogeneous information for path planning and intelligent obstacle avoidance during driving; A8: After the dispatch is completed, the communication module (26) feeds back the logistics dispatch information and starts the next round of cargo dispatch after returning to the cargo transfer point.

Images