Full-scene automatic driving positioning device and method for mobile charging equipment

Abstract

The invention provides a full-scene automatic driving positioning device and method for mobile charging equipment, and relates to the technical field of new energy automobile charging. The device comprises an automatic driving system, a map cloud platform in communication connection with the automatic driving system, an automatic driving algorithm training cloud platform, an interactive terminal man-machine interface and a plurality of positioning sensors arranged on mobile charging equipment. Through the automatic driving system with a sensing layer, a planning and decision-making layer and a control layer, and by means of a map cloud platform, an automatic driving algorithm training cloud platform and a plurality of positioning sensors, positioning can be provided for mobile charging equipment in all-scene parking areas such as indoor and outdoor, ground and underground, all-scene automatic driving is realized, and the driving efficiency is improved. Any new energy automobile in the parking area is charged, and flexible mobile charging of the new energy automobile is achieved.

Description

Technical Field

[0001] The present invention relates to the field of new energy vehicle charging technology, and in particular to a full-scenario automatic driving positioning device and method for a mobile charging device. Background Art

[0002] With the widespread popularity of new energy vehicles, many charging problems have arisen, such as insufficient new energy vehicle charging stations, long charging time, and long-term occupancy of charging stations. Therefore, the original fixed charging piles can no longer meet the needs of the new energy market for flexible and efficient charging. For this reason, new charging modes have emerged in the market, such as automatic charging equipment and mobile automatic charging equipment, which make it possible to achieve flexible, convenient, and efficient charging.

[0003] However, most of the current mobile charging equipment is in a semi-automatic state. For example, some mobile charging equipment needs to use special tracks to move within a set fixed route, which requires additional track laying. The investment in building a large area of ​​additional tracks is high, so there are fewer applicable scenarios and flexibility is still insufficient.

[0004] Based on this, the present invention is proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a mobile charging device full-scenario automatic driving positioning device and method, to provide positioning for the mobile charging device, to realize automatic driving in all scenarios such as indoors and outdoors, on the ground, underground, etc., and to realize flexible mobile charging of new energy vehicles.

[0006] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, the present invention provides a full-scenario automatic driving positioning device for a mobile charging device, comprising an automatic driving system, a map cloud platform connected to the automatic driving system for communication, an automatic driving algorithm training cloud platform, an interactive terminal human-machine interface, and a plurality of positioning sensors arranged on the mobile charging device; the automatic driving system comprises a perception layer, a planning and decision layer, and a control layer; wherein the perception layer performs data interaction with the automatic driving algorithm training platform and the map cloud platform, and is at least used to receive data from a plurality of positioning sensors to perceive the surrounding environment data of the mobile charging device and the mobile charging device-related data, obtain and fuse the data, and construct a surrounding environment map with the help of the map cloud platform; the planning and decision layer performs data interaction with the interactive terminal human-machine interface, and is at least used to perform path planning, driving behavior decision-making, motion planning, and send operation instructions for the mobile charging device according to the surrounding environment data and in combination with the surrounding environment map; the control layer is at least used to control the mobile charging device to perform corresponding actions according to the motion plan according to the operation instructions.

[0007] As a preferred solution of the first aspect, in order to adapt to the accuracy of positioning in various parking environments, the multiple positioning sensors include a GPS locator, a multi-beam laser radar, an inertial measurement unit (IMU) and a binocular depth camera; wherein the GPS locator is used to obtain traffic operation data; the multi-beam laser radar is used to obtain vehicle position data of the mobile charging device; the inertial measurement unit is used to obtain vehicle driving data of the mobile charging device; and the binocular depth camera is used to obtain vehicle environment data of the mobile charging device.

[0008] As a preferred scheme in the first aspect, the fused data is realized by the following method: using a Kalman filter, a particle filter or a graph optimization algorithm to perform data fusion processing on the vehicle position data of the mobile charging device, the surrounding vehicle position data, the pedestrian position data, the distance data of the surrounding environment, the surrounding environment image data, and the driving motion status data.

[0009] In a second aspect, the present invention provides a full-scenario automatic driving positioning method for a mobile charging device, characterized in that it is based on an automatic driving system, a map cloud platform communicated with the automatic driving system, an automatic driving algorithm training cloud platform, an interactive terminal human-machine interface and multiple positioning sensors arranged on the mobile charging device; the automatic driving system perception layer, planning and decision-making layer and control layer; the method includes: perception, data interaction with the automatic driving algorithm training platform and the map cloud platform, receiving data from multiple positioning sensors to perceive the surrounding environment data of the mobile charging device and the mobile charging device related data, acquire and fuse the data, and construct a surrounding environment map with the help of the map cloud platform; planning and decision-making, data interaction with the interactive terminal human-machine interface, path planning, driving behavior decision-making, motion planning and sending operation instructions for the mobile charging device according to the surrounding environment data and combined with the surrounding environment map; control, according to the operation instructions, control the mobile charging device to perform corresponding actions according to the motion plan.

[0010] Compared with the prior art, the above technical solution has the following advantages: The device and method for full-scenario automatic driving positioning of a mobile charging device of the present invention can provide positioning for the mobile charging device in full-scenario parking areas such as indoors and outdoors, on the ground, and underground, through an automatic driving system having a perception layer, a planning and decision-making layer, and a control layer, and with the help of a map cloud platform, an automatic driving algorithm training cloud platform, and multiple positioning sensors, thereby realizing full-scenario automatic driving, charging any new energy vehicle in the parking area, and realizing flexible mobile charging of new energy vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0012] Figure 1 This is a block diagram of a full-scenario automatic driving positioning device for a mobile charging device provided in Embodiment 1 of the present invention; Figure 2 A flow chart of a full-scenario automatic driving positioning method for a mobile charging device provided in Example 1 of the present invention; Figure 3 A block diagram of a full-scenario automatic driving positioning device for a mobile charging device provided in Embodiment 2 of the present invention; Figure 4 A strategy flow chart of a behavior decision module in a full-scenario autonomous driving positioning device for a mobile charging device provided in Example 2 of the present invention.

[0013] Reference numerals: autonomous driving system 100, perception layer 110, environment perception subsystem 111, surrounding vehicle data acquisition module 1111, pedestrian position data acquisition module 1112, road target perception data acquisition module 1113, traffic operation environment perception data acquisition module 1114, road infrastructure perception data acquisition module 1115, vehicle data acquisition module 1116, real-time vehicle environment perception map subsystem 112, scheduling operation data generation module 1121, road target data generation module 1122, road infrastructure data generation module 1123, high-precision map generation module 1124, planning and decision-making layer 120, path planning module 121, behavior decision-making module Block 122, motion planning module 123, operation instruction generation module 124, control layer 130, electronic control unit 131, interactive terminal human-machine interface 200, mobile charging device on-board terminal 210, vehicle terminal 220, roadside terminal 230, pedestrian terminal 240, autonomous driving algorithm training cloud platform 300, map cloud platform 400, scheduling management cloud platform 500, sensor one 610, GPS locator 611, sensor two 620, binocular depth camera 621, sensor three 630, multi-beam laser radar 631, sensor four 640, inertial measurement unit 641, sensor five 650, driving sensor 651, ultrasonic sensor 661, actuator 700. DETAILED DESCRIPTION

[0014] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0015] Example 1

[0016] Please refer to Figure 1 , a mobile charging device full-scenario automatic driving positioning device of this embodiment includes an automatic driving system 100, a map cloud platform 400 communicating with the automatic driving system 100, an automatic driving algorithm training cloud platform 300, an interactive terminal human-machine interface 200 and a plurality of positioning sensors arranged on the mobile charging device; the automatic driving system 100 includes a perception layer 110, a planning and decision-making layer 120 and a control layer 130; wherein, The perception layer 110 interacts with the autonomous driving algorithm training platform and the map cloud platform 400 to receive data from multiple positioning sensors (sensor 1 610, sensor 2 620, sensor 3 630, sensor 4 640, sensor 5 650) to perceive the surrounding environment data of the mobile charging device (including vehicle environment data, traffic operation data, etc.) and mobile charging device related data (vehicle location data, vehicle driving data, vehicle data, etc.), obtain and fuse the data, and build a surrounding environment map with the help of the map cloud platform 400. The map cloud platform 400 can help the perception layer 110 build a surrounding environment map based on the surrounding environment data and mobile charging device related data obtained by the perception layer 110. The surrounding environment data and mobile charging device related data obtained by the perception layer 110 are often massive data, which need to be efficiently stored and processed for subsequent analysis and algorithm training. The autonomous driving algorithm training cloud platform 300 provides powerful data storage and processing capabilities, which can cope with the huge amount of data generated by autonomous driving vehicles and ensure the integrity and accuracy of the data.

[0017] The planning and decision layer 120 interacts with the interactive terminal human-machine interface 200 to perform path planning, driving behavior decision-making, motion planning, and send operation instructions for the mobile charging device based on the surrounding environment data and in combination with the surrounding environment map. The planning and decision layer 120 includes a path planning module 121, a behavior decision module 122, a motion planning module 123, and an operation instruction generation module 124. In this embodiment, the user can input the charging requirements on the interactive terminal human-machine interface 200, and the planning and decision layer 120 performs path planning and behavior decisions based on the user's charging requirements, and feeds back the relevant path suggestions and driving suggestions to the user through the interactive terminal human-machine interface 200. In addition, the user can also input the operation intention, so that the control layer 130 controls the actuator 700 of the mobile charging device to execute according to the user's operation intention.

[0018] The control layer 130 is used to control the mobile charging device to perform corresponding actions according to the motion plan according to the operation instructions. The control layer 130 receives the driver's operation intention from the interactive terminal human-machine interface 200 and the operation instructions from the planning and decision-making layer 120, and controls the actuator 700 of the mobile charging device to perform corresponding operations according to the operation intention and the operation instruction, and performs a certain motion at a certain position and time point according to the planned path. At the same time, a driving sensor 651 is set at the actuator 700 to feedback vehicle data to the electronic control unit 131. Considering the ever-changing environment, the perception layer 110 uploads the surrounding environment data and the vehicle data of the mobile charging device in real time, and the operation instructions generated by the planning and decision-making layer 120 are often real-time. The control layer 130 of this embodiment adopts an electronic control unit 131 (ECU).

[0019] In order to ensure that the autonomous driving system 100 understands and recognizes the environment, the system's perception layer 110 needs to obtain a large amount of information about the surrounding environment. The perception layer 110 integrates this data and passes the correct information to the planning and decision-making module for use. The planning and decision-making layer 120 is one of the core modules of the autonomous driving system 100. It takes over the upstream perception layer 110, plans the specific driving behavior of the vehicle over a period of time, and outputs it to the downstream control module in the form of a trajectory for execution. The control layer 130 is the bottom layer of the autonomous driving system 100. Its task is to implement the planned actions, so the evaluation index of the control layer 130 is the accuracy of control. There will be measurements inside the control layer 130. The control layer 130 outputs control actions by comparing the vehicle's measurements with the expected state. This process is called feedback control.

[0020] Please refer to Figure 2The present embodiment provides a mobile charging device full-scenario autonomous driving positioning method, based on an autonomous driving system 100, a map cloud platform 400 communicating with the autonomous driving system 100, an autonomous driving algorithm training cloud platform 300, an interactive terminal human-machine interface 200, and a plurality of positioning sensors disposed on the mobile charging device; the autonomous driving system 100 perception layer 110, planning and decision layer 120, and control layer 130; the method includes: S110 perception, which interacts with the autonomous driving algorithm training platform and the map cloud platform 400, receives data from multiple positioning sensors, perceives the surrounding environment data of the mobile charging device and the mobile charging device related data, obtains and fuses the data, and constructs a surrounding environment map with the help of the map cloud platform 400; S120 planning and decision making, performing data interaction with the interactive terminal human-machine interface 200, performing path planning, driving behavior decision making, motion planning, and sending operation instructions for the mobile charging device based on the surrounding environment data and in combination with the surrounding environment map; S130 controls the mobile charging device to execute corresponding actions according to the motion plan according to the operation instruction.

[0021] The full-scenario automatic driving positioning device and method for mobile charging equipment in this embodiment, through the automatic driving system 100 having a perception layer 110, a planning and decision-making layer 120, and a control layer 130, and with the help of a map cloud platform 400, an automatic driving algorithm training cloud platform 300 and multiple positioning sensors, can provide positioning for mobile charging equipment in full-scenario parking areas such as indoors and outdoors, on the ground, and underground, realize full-scenario automatic driving, charge any new energy vehicle in the parking area, and realize flexible mobile charging of new energy vehicles.

[0022] Example 2

[0023] Based on Example 1, this example adds the following technical solutions: Please refer to Figure 3 The full-scenario automatic driving positioning device for mobile charging equipment of this embodiment is also provided with a dispatching management cloud platform 500 that is connected to the automatic driving system 100 in communication; the planning and decision layer 120 also interacts with the dispatching management cloud platform 500 for data, and also performs path planning, driving behavior decision-making, motion planning, and sending operation instructions for the mobile charging equipment in combination with the obtained dispatching-related data. That is, the planning and decision layer 120 of this embodiment also needs to combine the dispatching data sent by the dispatching management cloud platform 500 and the data uploaded by the perception layer 110 to make corresponding planning and decisions. Through the dispatching management cloud platform 500, this embodiment can manage and dispatch multiple mobile charging devices according to different parking scenes, which can fully mobilize resources, reduce operating costs, and improve economic benefits. At the same time, it also enables multiple mobile charging devices to provide services to users in an orderly manner under dispatching management.

[0024] The perception layer 110 of this embodiment includes two subsystems, namely, an environmental perception subsystem 111 and a real-time vehicle environment perception map subsystem 112. Among them, the environmental perception subsystem 111 obtains relevant data from multiple positioning sensors, which mainly includes a surrounding vehicle data module, a pedestrian position data acquisition module, a road target perception data acquisition module, a traffic operation environment perception data acquisition module, a road infrastructure perception data acquisition module, and a vehicle data acquisition module. The environmental perception subsystem 111 includes a surrounding vehicle data acquisition module 1111, a pedestrian position data acquisition module 1112, a road target perception data acquisition module 1113, a traffic operation environment perception data acquisition module 1114, a road infrastructure perception data acquisition module 1115, and a vehicle data acquisition module 1116. In this embodiment, the multiple positioning sensors mainly include the following: a GPS locator 611, a multi-beam laser radar, an inertial measurement unit 641, and a binocular depth camera 621; wherein, the GPS locator 611 is used to obtain traffic operation data, and obtains corresponding traffic operation data by positioning other surrounding vehicles. The binocular depth camera 621 is used to obtain the vehicle environment data of the mobile charging device, which is the image data of the surrounding environment. The binocular depth camera 621 calculates the position and depth information of the object in three-dimensional space through the parallax of the two cameras. Since the binocular system can capture more spatial information by utilizing the difference in viewing angles of the two cameras, it can provide high-precision depth measurement. This high precision gives the binocular depth camera 621 significant advantages in the fields of object size measurement, three-dimensional reconstruction, and robot navigation. The multi-beam laser radar is used to obtain the vehicle position data of the mobile charging device, which obtains the specific position by detecting the distance to other surrounding vehicles. The inertial measurement unit 641 is used to obtain the vehicle driving data of the mobile charging device and obtain the vehicle driving status in real time.

[0025] The real-time vehicle environment perception map subsystem 112 includes a scheduling operation data generation module 1121, a road target data generation module 1122, a road infrastructure data generation module 1123, and a high-precision map generation module 1124. The real-time vehicle environment perception map subsystem 112, with the help of the map cloud platform 400 and in combination with the data uploaded by the environment perception subsystem 111, can generate a high-precision map of the vehicle's surrounding environment, road target data, and road infrastructure data in real time, and with the help of the scheduling management cloud platform 500, generate scheduling operation data for the planning and decision-making layer 120 to plan and make decisions.

[0026] The Global Positioning System (GPS) is one of the most widely used positioning technologies. It determines the position of the receiver by receiving satellite signals. However, GPS signals are susceptible to building obstruction and electromagnetic interference. If only the GPS locator 611 is used, the positioning accuracy will be reduced. The inertial measurement unit 641 (IMU) estimates the motion state of an object by measuring acceleration and angular velocity. IMU positioning does not rely on external signals, but there is a cumulative error, so it is usually used in combination with other sensors. Laser radar (LiDAR) measures the precise distance of the surrounding environment by emitting laser pulses and receiving the reflected signals. LiDAR positioning has high accuracy, and its performance will be affected in severe weather conditions. The reason why this embodiment uses a multi-beam laser radar is that: as the simultaneous localization and mapping (SLAM) technology plays a vital role in the navigation of autonomous systems such as autonomous vehicles and robots. In recent years, with the development of laser radar technology, SLAM technology based on laser radar has attracted widespread attention due to its ability to maintain robustness to changes in illumination and obtain accurate distance information. Based on a single-line laser, it is called 2D-SLAM. Due to the number of feature points that a single-line laser can provide, it is easy to fail in positioning when the environment is complex and changing. Based on multi-beam laser, it is called 3D-SLAM. Multi-beam laser can extract many feature points and is suitable for complex and changing environments. The binocular depth camera 621 can capture image information of the surrounding environment, identify road signs, lane lines and other features through image processing technology, and realize visual positioning. Visual positioning is sensitive to environmental changes. Therefore, if only the above-mentioned single positioning sensor is adopted, the applicable scenes are limited. For example, if only a single laser is used, the calculation amount is large and it is easy to diverge. Therefore, in order to make the final positioning data more stable, it is necessary to fuse the above-mentioned multiple positioning sensors to further apply to full-scene autonomous driving positioning. R3LIVE++ is a lidar inertial vision fusion framework that combines two subsystems: LIO and VIO subsystems. These two subsystems jointly and incrementally build a 3D radiation map of the surrounding environment in real time. LIO reconstructs the geometric structure by aligning the new point cloud in LiDAR to the map, while VIO restores the radiation information by rendering the pixel color in the image to the point in the map; vision is limited by the environment, and the environment is dim and harsh, especially underground, which greatly reduces the visual effect.

[0027] The interactive terminal human-machine interface 200 of this embodiment includes a mobile charging device on-board terminal 210, a vehicle terminal 220 (vehicle terminal 220 of other vehicles), a roadside terminal 230 and a pedestrian terminal 240; wherein, the mobile charging device on-board terminal 210 is connected to the planning and decision-making layer 120 of the automatic driving system 100, and is used to obtain the user's mobile charging requirements and send them to the planning and decision-making layer 120, and the planning and decision-making layer 120 sends path suggestions and / or driving suggestions to the mobile charging device on-board terminal 210; the vehicle terminal 220, the roadside terminal 230 and the pedestrian terminal 240 exchange data with the perception layer 110 of the automatic driving system 100 through VZX, obtain surrounding vehicle data, road traffic data and pedestrian location data, and verify them with the data obtained by multiple sensors, so as to further improve the accuracy of positioning.

[0028] Example 3

[0029] Based on Example 2, this example differs from Example 2 in data fusion: The multi-sensor fusion technology provided in this embodiment combines the multiple sensor data of GPS locator 611, inertial measurement unit 641 (IMU), multi-beam laser radar (LiDAR), and binocular depth camera 621, and improves the accuracy and reliability of positioning through data fusion algorithm. The fusion data of this embodiment is realized by the following data fusion algorithm: the vehicle position data of the mobile charging device, the surrounding vehicle position data, the pedestrian position data, the distance data of the surrounding environment, the surrounding environment image data, and the driving motion state data are processed by data fusion using Kalman filter, particle filter or graph optimization algorithm. In this embodiment, the extended Kalman filter (EKF) is preferably used as the main data fusion means, combined with the high-frequency update of IMU and the global positioning information of GPS locator 611, as well as the environmental feature information of multi-beam laser radar and binocular depth camera 621. The specific process is as follows: Initialization: The initial position is obtained through the GPS locator 611, and the inertial measurement unit 641 is started to estimate the state at the same time. Prediction: The next state of the vehicle is predicted based on the data of the inertial measurement unit 641. Update: Use data from multi-beam laser radar and binocular depth camera 621 to correct the predicted state. Optimize: Use graph optimization technology to perform loop detection on historical data to further improve positioning accuracy.

[0030] This embodiment has been verified through experiments that in a ground parking environment, the positioning accuracy of the GPS locator 611 is high, but there is an occasional jump point phenomenon. In an underground parking environment, the fusion positioning of the inertial measurement unit 641 and the multi-beam lidar can effectively make up for the shortcomings of the GPS signal. The binocular depth camera 621, as a visual positioning sensor, provides additional auxiliary information in lane keeping and traffic sign recognition. The extended Kalman filter (EKF) can effectively fuse different sensor data to improve the robustness of positioning. The full-scene autonomous driving positioning significantly improves the accuracy and reliability of positioning through the multi-sensor fusion technology of this embodiment.

[0031] In a more preferred embodiment, the autonomous driving positioning device is also provided with an ultrasonic sensor 661 for detecting the distance to surrounding obstacles, further achieving obstacle avoidance, and ensuring that the vehicle of the mobile charging device is not damaged.

[0032] Example 4

[0033] Based on Example 2 or 3, this embodiment provides a driving behavior decision strategy: Please refer to Figure 4 In the planning and decision-making layer 120, the driving behavior decision-making strategy for encountering a stationary or moving obstacle includes the following: S1221 obtains the state of the obstacle ahead; S1222 determines the state of the obstacle; if the obstacle ahead is stationary, S1223 determines whether there is a safe bypass passage, and if there is a safe bypass passage, S1224 adopts a bypass strategy; if there is no bypass condition, S1225 emergency braking stops; if the front vehicle is moving, S1226 follows the vehicle.

[0034] A driving behavior decision strategy provided by the automatic driving positioning device of this embodiment can realize the automatic lane changing function for the vehicle of the mobile charging device, and can actively detect various obstacles in front of the vehicle, and realize automatic lane changing when an obstacle appears in front of the vehicle and the time distance between the obstacle and the obstacle meets the lane changing conditions, and the side road allows lane changing; or realize automatic lane changing under path planning, so that the vehicle of the mobile charging device can reach the parking position of the vehicle to be charged more efficiently, quickly and flexibly.

[0035] Example 5

[0036] Based on Embodiment 2, 3 or 4, this embodiment provides several path planning modes: In the planning and decision-making layer 120, path planning is used to plan the driving path for each mobile charging driving task, including one of the following modes: Mode 1, teaching mode, recording and saving the manual teaching operation driving route, and the entire section or any section of the route is used as a planned path. In this mode, the operator remotely controls the vehicle of the mobile charging device to walk along the working route, and the vehicle of the mobile charging device will record the route. The administrator can start the route saved after teaching according to the ID of the route, and can start at any point in the route. Mode 2, site line mode, set a number of fixed sites, connect each site to form a route as a planned path. In this mode, like a bus route, the operator fixes several sites in a task, and combines several sites to form a task driving similar to a bus route. The vehicle of the mobile charging device walks between the sites in sequence like a bus. Mode 3, optimal path mode, actively plans the optimal path on the map according to the target location as a planned path. In this mode, a certain location in the map calls the vehicle of the mobile charging device, and the vehicle of the mobile charging device plans the optimal path to reach the call location like a taxi. Mode 4, line drawing navigation mode, according to the path drawn manually on the map as a planned path. In this mode, the operator can draw a path on the map, and the vehicle of the mobile charging device can navigate along the drawn path. The above modes can also be combined arbitrarily. After executing one mode, it can immediately switch to another mode, and finally complete the task in the best way through the combination. This embodiment can meet the diverse needs of users and improve the user experience by setting multiple path planning modes.

[0037] In addition to the above embodiments, the automatic driving positioning device of the present invention is also provided with the following preferred functions: Adaptive lane keeping function: The system must be able to ensure that the vehicle does not deviate from its lane and that the vehicle remains in a relatively fixed lateral position in its lane to avoid disrupting normal traffic flow.

[0038] Obstacle avoidance maneuver within the lane: When a static obstacle is encountered suddenly (such as the vehicle in front avoiding the obstacle or an aggressive lane change), and the vehicle cannot avoid the collision by slowing down alone, it is necessary to perform obstacle avoidance maneuver within the lane.

[0039] Remote intervention: When the remote controller correctly performs the intervention, the system should safely and smoothly hand over control back to the human operator, who will then take control of the vehicle movement.

[0040] Adaptive speed keeping function: The system can adaptively adjust the vehicle speed according to the current traffic conditions.

[0041] Adaptive distance keeping function: The system can adaptively adjust the distance to the vehicle in front according to the current traffic conditions, and this distance needs to be greater than the minimum following distance.

[0042] System activation and deactivation: The system should provide the driver with a dedicated activation and deactivation device to prevent accidental triggering.

[0043] Able to display the operating status of the system: The system must be able to display its status, such as normal operation, fault, lateral override, etc.

[0044] Driver takeover request: The driver can be notified in an appropriate manner when driver takeover is required.

[0045] Able to perform emergency braking when necessary: ​​When encountering situations such as a vehicle in the side lane urgently cutting in or the vehicle in front urgently brakes, the system must be able to perform emergency braking to avoid a collision.

[0046] The system can respond to pause, stop, speed limit and other instructions from the dispatching system and interact with the management center.

[0047] Fixed route driving: The system should be able to complete the automatic driving task of a fixed route according to the fixed route specified by the customer; the fixed route includes the starting point, end point, parking point, etc. Fixed-point parking function: The system can rely on the vehicle's own sensors to realize parking at a designated location on the driving path.

[0048] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the above-mentioned embodiments of the methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0049] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0050] The above-described embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

Claims

1. A full-scenario automatic driving positioning device for mobile charging equipment, characterized in that: It includes an autonomous driving system, a map cloud platform connected to the autonomous driving system, an autonomous driving algorithm training cloud platform, an interactive terminal human-machine interface, and multiple positioning sensors arranged on a mobile charging device; the autonomous driving system includes a perception layer, a planning and decision-making layer, and a control layer; wherein, A perception layer, which interacts with the autonomous driving algorithm training platform and the map cloud platform, and is at least used to receive data from multiple positioning sensors to perceive the surrounding environment data and mobile charging device related data of the mobile charging device, obtain and fuse the data, and build a surrounding environment map with the help of the map cloud platform; The planning and decision-making layer performs data interaction with the interactive terminal human-machine interface, and is at least used to perform path planning, driving behavior decision-making, motion planning, and sending operation instructions for the mobile charging device based on the surrounding environment data and in combination with the surrounding environment map; The control layer is at least used to control the mobile charging device to perform corresponding actions according to the motion plan based on the operation instructions.

2. The full-scenario automatic driving positioning device for mobile charging equipment according to claim 1 is characterized in that: It also includes a scheduling management cloud platform that is communicatively connected to the autonomous driving system; the planning and decision-making layer also interacts with the scheduling management cloud platform for data, and also performs path planning, driving behavior decision-making, motion planning, and sending operation instructions for the mobile charging equipment in combination with the acquired scheduling-related data.

3. The full-scenario automatic driving positioning device for mobile charging equipment according to claim 1 is characterized in that: The interactive terminal human-machine interface includes a mobile charging device on-board terminal, a vehicle terminal, a roadside terminal and a pedestrian terminal; wherein the mobile charging device on-board terminal is connected to the planning and decision-making layer of the automatic driving system, and is used to obtain the user's mobile charging requirements and send them to the planning and decision-making layer, and the planning and decision-making layer sends path suggestions and / or driving suggestions to the mobile charging device on-board terminal; the vehicle terminal, the roadside terminal and the pedestrian terminal interact with the perception layer of the automatic driving system to obtain surrounding vehicle data and pedestrian location data.

4. The full-scenario automatic driving positioning device for mobile charging equipment according to claim 1 is characterized in that: The multiple positioning sensors include a GPS locator, a multi-beam laser radar, an inertial measurement unit and a binocular depth camera; wherein the GPS locator is used to obtain traffic operation data; the multi-beam laser radar is used to obtain vehicle position data of the mobile charging device; the inertial measurement unit is used to obtain vehicle driving data of the mobile charging device; and the binocular depth camera is used to obtain vehicle environment data of the mobile charging device.

5. The full-scenario automatic driving positioning device for mobile charging equipment according to claim 4 is characterized in that: The fusion data is realized by the following method: using Kalman filter, particle filter or graph optimization algorithm to perform data fusion processing on the vehicle position data of the mobile charging device, the surrounding vehicle position data, the pedestrian position data, the distance data of the surrounding environment, the surrounding environment image data, and the driving motion state data.

6. The full-scenario automatic driving positioning device for mobile charging equipment according to claim 4 is characterized in that: In the fused data, the data acquired by the multi-beam LiDAR and the inertial measurement unit are fused using the R3LIVE++ fusion framework with the LIO subsystem and the VIO subsystem; the LIO subsystem and the VIO subsystem jointly and incrementally construct a 3D radiation map of the surrounding environment in real time. The LIO reconstructs the geometric structure by aligning the new point cloud in the LiDAR to the map, while the VIO restores the radiation information by rendering the pixel color in the image to the point in the map.

7. The full-scenario automatic driving positioning device for mobile charging equipment according to claim 4 is characterized in that: In the planning and decision-making layer, the driving behavior decision-making includes the driving behavior decision-making strategy when encountering a stationary or moving obstacle, including the following: if the obstacle in front is stationary, if there is a safe bypass passage, then adopt a bypass strategy; if there is no bypass condition, then emergency braking and stopping are performed; if the vehicle in front is moving slowly, then follow the vehicle.

8. The full-scenario automatic driving positioning device for mobile charging equipment according to claim 4 is characterized in that: The plurality of positioning sensors further include an ultrasonic sensor for detecting the distance from surrounding obstacles.

9. The full-scenario automatic driving positioning device for mobile charging equipment according to claim 1, characterized in that: In the planning and decision-making layer, path planning is used to plan a driving path for each mobile charging driving task, including one of the following modes: Teaching mode, record and save the manual teaching operation route, and use the whole route or any section as the planned path; The station-line mode sets up a number of fixed stations and connects them to form a route as a planned path; The optimal path mode actively plans the optimal path on the map according to the target location as the planned path; Line drawing navigation mode, the path drawn manually on the map is used as the planned path.

10. A full-scenario automatic driving positioning method for a mobile charging device, characterized in that: Based on an autonomous driving system, a map cloud platform connected to the autonomous driving system, an autonomous driving algorithm training cloud platform, an interactive terminal human-machine interface, and multiple positioning sensors arranged on a mobile charging device; the autonomous driving system perception layer, planning and decision layer, and control layer; the method includes: Perception: interacting with the autonomous driving algorithm training platform and the map cloud platform, receiving data from multiple positioning sensors to perceive the surrounding environment data and mobile charging device related data of the mobile charging device, acquiring and fusing the data, and building a surrounding environment map with the help of the map cloud platform; Planning and decision-making: data interaction with the interactive terminal human-machine interface, path planning, driving behavior decision-making, motion planning and sending operation instructions for the mobile charging device based on the surrounding environment data and combined with the surrounding environment map; Control, according to the operation instructions, control the mobile charging device to perform corresponding actions according to the motion plan.

Images