A vertical empty container area operation method through unmanned container truck self-perception

Abstract

The application discloses a vertical empty container area operation method through self-vehicle sensing of an unmanned truck, wherein the unmanned truck stops at a coordinate point after entering the container area, starts laser container line sensing alignment function, starts to identify the longitudinal distance from the container through self-vehicle sensing, stops after identifying the container line of the first container on the outermost side, and performs container grabbing or unloading operation by the stacker. After the operation is completed, the unmanned truck confirms the loading or unloading state through self-vehicle sensing, combines the sensing surrounding condition, and safely drives away from the operation position. The stacker in the non-standard empty container area performs operation in the order from the entry gate to the exit gate, solves the problem that the unmanned truck cannot stop at the position through the corresponding relationship between the berth number and the container in the non-standard empty container area. Meanwhile, the unmanned truck stops at the position through self-vehicle sensing and identification of the container line of the container on the outermost side, and the accuracy of stopping is improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of automatic driving, and particularly relates to a vertical empty container area operation method through self-perception of an unmanned container truck. BACKGROUND

[0002] With the continuous expansion of port business, port automation and unmanned operation have become an inevitable trend to reduce costs and improve port operation efficiency. With the help of wireless communication, cloud computing, map positioning, vehicle-end sensing and identification technologies, unmanned container trucks (i.e., unmanned container trucks) can perform automatic loading and unloading operations in scenarios such as port gantry crane areas, shore crane areas, and empty container areas. Heavy containers need to be loaded and unloaded by large machinery such as gantry cranes, while empty containers can be operated by stackers with higher flexibility and lower prices. Therefore, heavy containers and empty containers are generally stacked separately in the gantry crane area and the empty container area, and the gantry crane and the stacker operate separately to improve the transfer efficiency and cost of the port.

[0003] In the existing scenario, the empty container area for unmanned container truck operation is a standard empty container area, the ground is marked with a berth number, and the containers are placed in order according to the berth number, with the container bodies parallel to the operation lane. The unmanned container truck stops by the berth number.

[0004] However, due to limited port space and increasing port throughput, some ports stack containers beyond the planned stacking area or change from horizontal placement of containers to vertical placement, i.e., the container bodies are perpendicular to the operation lane, and the original berth numbers cannot correspond to the containers. This non-standard vertical empty container area scenario increases the difficulty of parking for unmanned container trucks.

[0005] Meanwhile, in the prior art, after the stacker and the unmanned container truck interact with each other, the stacker driver and the driver or the following safety officer of the unmanned container truck need to use gestures or other forms of gestures to indicate, and the driver or the safety officer of the unmanned container truck needs to manually intervene before driving away from the operation position. Alternatively, after the stacker is intelligently modified and performs operation, it sends a completed operation signal to the terminal operating system (TOS), and the TOS sends a signal to the unmanned container truck to drive away. However, the intelligent modification of the stacker requires time and cost, and some ports do not have such conditions, so manual intervention is still needed to confirm the completion of the operation, and the unmanned container truck needs to be equipped with a driver or a safety officer, which increases the labor cost.

[0006] The information disclosed in this BACKGROUND section is only for the purpose of increasing the understanding of the background of the present application and should not be regarded as an acknowledgment or any form of suggestion that it forms prior art of this application. SUMMARY

[0007] The present application aims to provide a vertical empty container area operation method through self-perception of an unmanned container truck, so as to overcome the defects in the prior art.

[0008] To achieve the above-mentioned purpose, the present application provides a vertical empty container area operation system through self-perception of an unmanned container truck, wherein the empty container area is a single lane, the entrance of the empty container area is an entrance, and the exit of the empty container area is an exit; the stacker starts to grab or drop containers from the position closest to the entrance in sequence during operation; characterized in that: a first container line corresponding to the first container position on the outermost side of the exit is provided, the first container line is an extension line of the first container body obtained by processing laser point cloud linear fitting; the unmanned container truck is provided with a perception module for perceiving the first container line; the unmanned container truck is also provided with a positioning module for obtaining the position of the vehicle, and a communication module for receiving and transmitting tasks from the port operation system.

[0009] Preferably, in the above technical solution, the positioning module is a module for realizing self-positioning through the fusion of global navigation satellite system, real-time dynamic carrier phase difference technology, inertial navigation and other technologies. The self-positioning information is obtained by processing and fusing the data obtained by the self-sensor of the vehicle.

[0010] The perception module is a module for obtaining external visual information by fusing environmental data through the combination of multiple sensors such as cameras, millimeter wave radars or laser radars; the positioning module enables the unmanned container truck to obtain real-time position information of itself; the perception module enables the unmanned container truck to stop at the parking position through real-time perception, and to confirm the loading and unloading state after completing the operation, and to drive away from the operation position by combining the perception of the surrounding conditions; the self-sensor of the vehicle can identify and distinguish the operation state of the stacker, such as the stacker with grabbed containers and the stacker without grabbed containers, and can judge the loading and unloading state by perceiving whether there are containers on the vehicle. For example, during the loading process, the stacker needs to grab the containers from the unmanned container truck and then transfer the containers to the container area; the self-sensor of the vehicle identifies that the stacker has grabbed the containers and has moved away from the vehicle, and perceives that there are no containers on the vehicle, so it is judged that the interactive operation has been completed. The same applies to the unloading process. The communication module is used for real-time information interaction between the unmanned container truck and the port operation system, receiving and transmitting task information, and accepting port scheduling.

[0011] After judging that the operation task has been completed and there is no risk of collision with other obstacles, the vehicle safely drives away from the operation position.

[0012] A vertical empty container area operation method through self-perception of an unmanned container truck, characterized by comprising the following steps:

[0013] S1. The unmanned container truck receives the task through the communication module, and stops at the obtained parking position after entering the empty container area;

[0014] S2, the unmanned container truck starts again after opening the laser box line sensing positioning function at the parking point, identifies the longitudinal distance of the first container extension box line through the self-vehicle sensing, stops after identifying the box line of the first container, and opens the laser after entering the box area; the risk of stopping after false detection outside the box area can be reduced, which refers to that if the above laser recognition function is opened in the box area, false detection may occur, that is, the box line of the container in other box areas instead of the target box area is recognized and the truck stops. That is, the truck stops before reaching the target position.

[0015] S3, after the unmanned container truck stops at the position of the first container closest to the entrance, the self-vehicle sensing completes the interaction with the stacker, the unmanned container truck exchanges real-time information with the terminal operating system (TOS) through the communication module, transmits and receives task information, and accepts the port scheduling. After determining that the task is completed and there is no collision risk with surrounding vehicles or other obstacles, the unmanned container truck safely drives away from the working position.

[0016] S4, after the unmanned container truck completes the task, if no new task is received within a certain time, for example, x minutes (x is a calibrated value, which can be changed according to the port demand), the unmanned container truck actively sends a task request combined with the state of the vehicle;

[0017] S5, if there is still no response, the unmanned container truck plans a path to drive away from the empty box area, which does not affect the subsequent container loading operation of other unmanned container trucks. If no new task is received after completing the task at the working position, the unmanned container truck will be parked for a long time, which will affect the subsequent container loading and unloading operation of other container trucks (not necessarily unmanned container trucks, but also ordinary container trucks).

[0018] Preferably, in the above technical solution, the specific steps of sensing the box line are:

[0019] The specific steps of sensing the box line are:

[0020] Point cloud data preprocessing;

[0021] Mapping three-dimensional point cloud information to a two-dimensional grid map;

[0022] Obtaining the approximate position of the empty box area relative to the unmanned container truck;

[0023] Searching for grid occupancy points in a certain area;

[0024] Linear fitting is performed on the searched grid occupancy points;

[0025] According to the characteristics of the container, the linear fitting container position is selected;

[0026] The distance between the self-vehicle and the container position is calculated;

[0027] The first container extension box closest to the entry column is parked within a distance x m, x is a markable value, which can be changed according to the port demand, and m is a distance unit meter.

[0028] In detail, the point cloud data and the two-dimensional grid map search the grid occupation points in a certain area range, remove the ground points, and retain the grid information of the obstacles. According to the characteristics of the container, the straight line container grid points are fitted by using the RANSAC or least square method. The relative position of the fitted straight line is calculated to obtain two mutually perpendicular straight lines. The intersection of the two straight lines is the corner point of the container, and the position of the corner point relative to the vehicle is calculated, that is, the position of the container relative to the vehicle.

[0029] Preferably, in the technical scheme, in step S1, if there are multiple container trucks at the entry column, the unmanned container truck identifies and queues for parking in combination with the sensing knowledge, and maintains a safe distance from the front vehicle. If there are container trucks queuing for parking at the entry before entering the container area, the unmanned container truck cannot enter, and the unmanned container truck waits for parking and follows the vehicle according to the distance perceived from the front vehicle.

[0030] Preferably, in the technical scheme, in step S3, the vehicle can identify and distinguish the working state of the stacker, such as the stacker having grabbed the container and the stacker not having grabbed the container, in combination with the sensor of the vehicle to perceive whether there is a container on the vehicle, so as to judge the loading and unloading state.

[0031] Compared with the prior art, the present application has the following beneficial effects:

[0032] In the disclosed scheme, the stacker in the non-standard empty container area works in sequence from the entry column to the exit column, solving the problem that the unmanned container truck cannot be parked by the berth number and the corresponding relationship of the container in the non-standard empty container area. At the same time, the outermost container line is identified by the vehicle sensing to realize autonomous parking, improving the accuracy of parking. Finally, the unmanned container truck realizes autonomous driving off through vehicle sensing, providing a feasible and safe solution for the port that has not been intelligently transformed, and the driver or safety officer of the unmanned container truck is no longer necessary, which can reduce labor costs and achieve truly efficient and low-cost unmanned driving. BRIEF DESCRIPTION OF DRAWINGS

[0033] Figure 1a Fig. a is a schematic diagram of the unmanned container truck sensing in the port empty container area in the embodiment of the present application;

[0034] Figure 1b Fig. b is a schematic diagram of the unmanned container truck sensing in the port empty container area in the embodiment of the present application;

[0035] Figure 2Flow chart for port empty container area operation of the unmanned straddle carrier in the embodiment of the present application through self-vehicle sensing;

[0036] Figure 3 Flow chart for the self-vehicle sensing parking function of the unmanned straddle carrier in the embodiment of the present application;

[0037] Figure 4 Flow chart for the operation after parking of the unmanned straddle carrier in the embodiment of the present application and driving away after the operation is completed. DETAILED DESCRIPTION

[0038] The specific embodiments of the present application are described in detail below, but it should be understood that the protection scope of the present application is not limited by the specific embodiments.

[0039] Unless otherwise explicitly stated, throughout the specification and claims, the term "comprise" or its variants such as "contain" or "include" and the like will be understood to include the stated element or component, but not to exclude the presence of other elements or components.

[0040] In order to solve the above problems, in order to provide an efficient and safe solution for the increasingly tight container stacking space in the port empty container area, the present application proposes an empty container area operation method through self-vehicle sensing of the unmanned straddle carrier for non-standard empty container area.

[0041] The scheme includes the container loading and unloading sequence in the port empty container area, from the entry gate to the exit gate of the container area, the straddle carrier starts to grab or place containers from the position closest to the entry gate in sequence when the straddle carrier is operated. The unmanned straddle carrier in the scheme at least includes a positioning module, a sensing module and a communication module. The unmanned straddle carrier obtains the real-time self-vehicle position through the positioning module, and realizes the parking at the coordinate point in the target container area in combination with the high-precision map. The focus of the present scheme is to park at the coordinate point after entering the container area, to start the laser container line sensing alignment function, to start identifying the longitudinal distance from the container through self-vehicle sensing, to stop the vehicle after identifying the container line of the first container on the outermost side, and to perform the container grabbing or unloading operation by the straddle carrier. After the operation is completed, the unmanned straddle carrier confirms the container loading and unloading state through self-vehicle sensing, and in combination with the sensing of the surrounding conditions, safely drives away from the operation position. Through the communication module, the unmanned straddle carrier and the terminal operating system (TOS) perform information interaction, and obtain operation task information and the like.

[0042] Specifically, the unmanned container truck is at least provided with a positioning module, a sensing module and a communication module. The positioning module is a module for realizing self-vehicle positioning through the fusion of global navigation satellite system, real-time dynamic carrier phase difference technology and inertial navigation technology. The sensing module is a module for obtaining external visual information by fusing environmental data through the combination of multiple sensors such as cameras, millimeter wave radars or laser radars. The positioning module enables the unmanned container truck to obtain real-time self-position information, and the sensing module enables the unmanned container truck to stop at a parking position through real-time sensing, confirm the loading and unloading state after completing the work, and drive away from the working position in combination with the sensing of the surrounding conditions. In addition to the positioning module and the sensing module, the unmanned container truck should at least include a communication module. Through the communication module, the unmanned container truck can interact with the terminal operating system in real time, receive and send task information, and accept port scheduling. After determining that the work task has been completed and there is no collision risk with other obstacles, the unmanned container truck drives away from the working position.

[0043] The application embodiment provides an unmanned container truck parking method in a port empty container area.

[0044] (1) The unmanned container truck receives a task, drives into the empty container area and stops at a parking position obtained after driving into the empty container area; if there are multiple container trucks at the entrance, the unmanned container truck identifies and queues to stop at a safe distance from the front vehicle.

[0045] (2) The unmanned container truck starts again after starting the laser box line sensing alignment function at the parking position, identifies the longitudinal distance of the container through self-sensing, stops after identifying the box line of the first container; the unmanned container truck can reduce the risk of stopping after false detection outside the container area.

[0046] The specific steps of sensing the box line are as follows:

[0047] 1. Point cloud data preprocessing;

[0048] 2. Mapping three-dimensional point cloud information to a two-dimensional grid map;

[0049] 3. Obtaining the approximate position of the empty container area relative to the unmanned container truck (left side or right side of the unmanned container truck);

[0050] 4. Searching for grid occupancy points in a certain area;

[0051] 5. Straight line fitting is performed on the searched grid occupancy points;

[0052] 6. Filtering the straight line fitting the container position according to the container characteristics;

[0053] 7. Calculate the distance from the truck to the container position;

[0054] Park within the first container extension box line xm (x is a scalable value, which can be changed according to the needs of the port, and m is the distance unit meter) closest to the entrance.

[0055] In detail, the point cloud data and the two-dimensional grid map are to search for grid occupation points within a certain area, remove ground points, and retain grid information of obstacles. According to the characteristics of the container, the straight line of the container position is fitted by using RANSAC or least squares method. The characteristics of the container grid points are two mutually perpendicular straight lines. Then, the relative position of the fitted straight line is calculated to obtain two mutually perpendicular straight lines. The intersection of the two straight lines is the corner point of the container, and the position of the corner point relative to the truck is calculated, that is, the relative position of the container to the truck. Specifically, after the laser box line sensing and positioning function of the unmanned truck is started, the three-dimensional point cloud information is mapped to the two-dimensional grid map according to the point cloud data obtained by the laser, and the approximate position of the container relative to the unmanned truck (left side or right side) is obtained. After obtaining the approximate position, the grid occupation points are searched within a certain area, the ground points are removed, and the grid information of the container obstacle is retained. According to the characteristics of the container, the straight line of the container position is fitted by using RANSAC or least squares method. The characteristics of the container grid points are two mutually perpendicular straight lines. Then, the relative position of the fitted straight line is calculated to obtain two mutually perpendicular straight lines. The intersection of the two straight lines is the corner point of the container, and the position of the corner point relative to the truck is calculated, that is, the relative position of the container to the truck. The distance of the first container closest to the entrance is extended by xm (x is a scalable value, which can be changed according to the needs of the port, and m is the distance unit meter) to park.

[0056] (3) After the unmanned truck stops at the position of the first container closest to the entrance, the truck senses the loading and unloading state; the truck can identify and distinguish the working state of the stacker through the sensing module, such as the stacker that has grabbed the container and the stacker that has not grabbed the container, and combine the sensing of whether there is a container on the truck to judge the loading and unloading state. For example, during the loading process, the stacker needs to grab the container from the unmanned truck and then transfer it to the container area. The truck identifies that the stacker has grabbed the container and has moved away from the truck, and senses that there is no container on the truck, so it is judged that the interactive work has been completed. The unloading process is the same. In the task, if the sensing identifies that the stacker stops near the unmanned truck with a longitudinal distance deviation, the truck adjusts the parking position according to the position of the stacker.

[0057] After the truck senses that the interactive work with the stacker has been completed, it sends a completed task signal to the terminal operating system (TOS) of the port, and after receiving the next task, the unmanned truck drives away from the empty container area in combination with the sensing of the surrounding situation.

[0058] (4) The unmanned container truck sends a task request actively after completing the task and exceeding x minutes (x is a calibratable value, which can be changed according to the port demand) without receiving the next task, and plans a path to leave the empty container area autonomously without affecting the subsequent container loading operation of the unmanned container truck.

[0059] The foregoing description of specific exemplary embodiments of the application has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the application to the precise forms disclosed, and obviously many modifications and variations are possible in light of the above teaching. It is intended that the scope of the application be limited not with this detailed description, but rather by the claims appended hereto.

[0060] The foregoing description of specific exemplary embodiments of the application has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the application to the precise forms disclosed, and obviously many modifications and variations are possible in light of the above teaching. It is intended that the scope of the application be limited not with this detailed description, but rather by the claims appended hereto.

Claims

1. A method for operating in a vertical empty container area using autonomous sensing by an unmanned container truck, characterized in that: The system includes a vertical empty container area operation system that uses an unmanned truck for sensing. The empty container area is a one-way street, with an entrance called an inlet and an exit called an outlet. When the forklift is operating, it grabs or places containers sequentially starting from the position closest to the inlet. The system is characterized by a first container line located at the position of the first container at the outlet, which is an extension line of the first container body obtained by fitting a straight line from a laser point cloud. The unmanned truck is equipped with a sensing module that detects the first container line. The unmanned truck also has a positioning module for acquiring the vehicle's location and a communication module for transmitting tasks from the terminal operating system. The positioning module is a module that integrates multiple technologies, including the Global Navigation Satellite System, real-time dynamic carrier phase differential technology, and inertial navigation, to achieve vehicle positioning. It processes and fuses data acquired by the vehicle's sensors to obtain the vehicle's position information. The perception module is a module that combines multiple sensors, such as cameras, millimeter-wave radar, or lidar, to fuse environmental data and obtain external visual information. The positioning module enables the unmanned truck to obtain its own position information in real time, while the perception module allows the unmanned truck to stop in real-time through perception when in a parking position, and to confirm the loading / unloading status through vehicle perception after completing operations, and then leave the work position based on perception of the surrounding environment. The communication module is used for real-time information exchange between the unmanned truck and the terminal operating system, sending and receiving task information, and accepting port dispatch. The work method is as follows: S1, the unmanned container truck receives the task through the communication module, drives into the empty container area and stops at the obtained parking point; S2: After the unmanned truck activates the laser container line sensing and alignment function at the parking point, it starts again. It identifies the longitudinal distance of the extended container line of the first container through its own vehicle sensing and stops after identifying the container line of the first container. The laser is activated again after the unmanned truck enters the container area. In S3, after the unmanned truck stops at the first container position closest to the gate, it senses that it has completed the interaction with the forklift. The unmanned truck then communicates with the terminal operating system (TOS) in real time through the communication module, sends and receives task information, and accepts port scheduling. If it determines that the task has been completed and there are no oncoming vehicles or other obstacles that pose a collision risk, it will safely leave the work position. In S3, the unmanned truck can identify and distinguish the working status of the forklift through the perception module, distinguishing between forklifts that have grabbed containers and those that have not. Combined with the truck's sensors, it can sense whether there are containers on the truck, thereby determining the loading and unloading status. S4: If the unmanned truck does not receive the next task after completing the task for a certain period of time, it will actively send a task request based on its own vehicle status. If there is still no response, S5 will autonomously plan its own route to leave the empty container area.

2. The method for operating in a vertical empty container area using autonomous sensing by an unmanned container truck as described in claim 1, characterized in that: The specific steps of the sensing box line are as follows: Point cloud data preprocessing; Mapping 3D point cloud information onto a 2D raster map; Obtain the approximate location of the empty container area relative to unmanned trucks; Search for grid occupancy points within a certain area; Perform a straight line fit on the searched grid occupancy points; Based on the characteristics of the containers, a straight line is selected to fit the position of the containers. Calculate the distance between the vehicle and the container. Park within x meters of the first container extension line closest to the entrance, where x is a calibrable value that can be changed according to port requirements, and m is the distance unit in meters.

3. The method for operating in a vertical empty container area using autonomous sensing by an unmanned container truck as described in claim 1, characterized in that: In step S1, if there are multiple container trucks at the entrance, the unmanned container trucks will use their own perception and recognition to queue up and park, maintaining a safe distance from the vehicle in front.

Images