Breaking-out method and control system of intelligent car

Abstract

The invention provides an intelligent car escape method and a control system. The method comprises the following steps: acquiring a current pause position of the intelligent car; determining a current grid where the intelligent car is located in a driving electronic map and a target position of a target two-dimensional code on an original navigation path based on the current position; and based on the position relationship between the current grid in the driving electronic map and the original navigation path, in combination with the current position and the attribute information of the grid on the original navigation path, determining an escape strategy for the intelligent car to arrive at the target position. The driving electronic map is used for compensating and assisting the intelligent car in getting out of the trouble, the navigation path can still be accurately returned from the trouble even in a complex environment, the getting-out difficulty is reduced, and the getting-out efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of navigation technology, in particular to a method for helping an intelligent small vehicle to get out of trouble and a control system. BACKGROUND

[0002] Intelligent Guided Vehicle (IGV) is an unmanned transport device used in automated ports and smart factories. It realizes autonomous navigation and obstacle avoidance through Beidou navigation system, laser radar, visual SLAM technology, etc. It has high flexibility, intelligent navigation and excellent performance.

[0003] At present, matrix two-dimensional codes are scattered and distributed, and the intelligent small vehicle obtains navigation information of the next matrix two-dimensional code under the driving value of a matrix two-dimensional code. Navigation information cannot be obtained between two matrix two-dimensional codes. When the intelligent small vehicle walks according to the navigation information of the matrix two-dimensional code, once a pause instruction is obtained, it must stay on a matrix two-dimensional code, so as to continue walking to the next matrix two-dimensional code according to the navigation information on the matrix two-dimensional code on which it stays when it starts again. If the IGV stays on a non-matrix two-dimensional code due to some reason, in order to enable the IGV to restart, it needs to return to any matrix two-dimensional code node on the navigation route by manual operation to get out of trouble.

[0004] Therefore, the present application provides a method for helping an intelligent small vehicle to get out of trouble to solve the above technical problems. SUMMARY

[0005] The present application aims to provide a method for helping an intelligent small vehicle to get out of trouble and a control system, which can solve at least one of the above technical problems. The specific scheme is as follows: According to the specific embodiment of the present application, in a first aspect, the present application provides a method for helping an intelligent small vehicle to get out of trouble, comprising: obtaining a current position where the intelligent small vehicle pauses; determining a current grid where the intelligent small vehicle is located in a driving electronic map and a target position of a target two-dimensional code on an original navigation path based on the current position, wherein the target two-dimensional code is a matrix two-dimensional code that the intelligent small vehicle has not traveled to in the original navigation path; determining a getting-out-of-trouble strategy of the intelligent small vehicle to reach the target position based on a position relationship between the current grid and the original navigation path in the driving electronic map, in combination with the current position and attribute information of the grid on the original navigation path.

[0006] Optionally, the determination of the getting-out-of-trouble strategy of the intelligent small vehicle to reach the target position comprises: determining a transfer grid on the original navigation path when the current grid is not a grid in the original navigation path in the driving electronic map; determining a strategy for the intelligent vehicle to reach the target position based on the current position, the transfer grid and a target grid where the target position is located, wherein the strategy comprises: the intelligent vehicle driving from the current position to the transfer grid, and then driving through corresponding grids based on recommended deviation coefficients and normal driving values of each grid on the original navigation path from the transfer grid, until reaching the target position.

[0007] Optionally, the determining a transfer grid on the original navigation path comprises: determining a grid closest to the current position in the original navigation path as the transfer grid when a recommended deviation coefficient of the grid meets a preset coefficient screening condition.

[0008] Optionally, the determining a grid closest to the current position in the original navigation path as the transfer grid when a recommended deviation coefficient of the grid meets a preset coefficient screening condition comprises: determining a grid closest to the current position in the original navigation path based on the current position; determining an original navigation section from the grid closest to the current position to a target grid in the original navigation path; determining a plurality of grids in front of the grid closest to the current position by a preset number of grids based on a driving direction in the original navigation section; screening at least one candidate grid with a minimum recommended deviation coefficient from the plurality of grids and the grid closest to the current position; determining a candidate grid closest to the current position from the at least one candidate grid as the transfer grid.

[0009] Optionally, the determining a strategy for the intelligent vehicle to reach the target position based on the current position, the transfer grid and a target grid where the target position is located comprises: determining a first strategy based on the current position, a normal start position of the transfer grid and the target position, wherein the first strategy comprises: the intelligent vehicle driving from the current position to the normal start position of the transfer grid, and then driving through corresponding grids based on recommended deviation coefficients and normal driving values of each grid on the original navigation path from the normal start position of the transfer grid, until reaching the target position.

[0010] Optionally, the determining a first strategy based on the current position, a normal start position of the transfer grid and the target position comprises: determining a transfer road section in the first escape strategy based on the current position and a normal start position of the transfer grid; determining normal start pose information and normal end pose information of each key grid passed by the transfer road section in the driving electronic map, and the normal start pose information of each key grid being consistent with the normal end pose information of a previous immediately adjacent key grid; taking the normal driving value and the recommended deviation coefficient in the transfer grid as the normal driving value and the recommended deviation coefficient of each key grid in the first escape strategy, wherein the normal driving value comprises a normal linear velocity value of each wheel and a normal angular velocity value of a corresponding wheel, and the recommended deviation coefficient comprises a recommended deviation coefficient of each wheel, and the first escape strategy comprises: the intelligent vehicle driving through a corresponding grid based on the recommended deviation coefficient and the normal driving value of each key grid passed on the transfer road section, starting from the current position, until reaching the normal start position of the transfer grid, and then driving through a corresponding grid based on the recommended deviation coefficient and the normal driving value of each grid passed on the original navigation path, starting from the normal start position of the transfer grid, until reaching the target position.

[0011] Optionally, the first escape strategy further comprises: triggering the pose verification of the intelligent vehicle once per key grid, wherein the pose verification of the intelligent vehicle comprises: when driving through a key grid, if the actual pose information of the intelligent vehicle satisfies a preset similar pose condition, then driving into a next key grid adjacent to the key grid; when driving through a key grid, if the actual pose information of the intelligent vehicle does not satisfy the preset similar pose condition, then correcting the actual pose information based on the normal end pose information of the key grid before driving into a next key grid adjacent to the key grid.

[0012] Optionally, the preset similar pose condition comprises: a deviation value between the actual position and the normal end position being less than or equal to 3 cm; and, a deviation value between the actual heading angle and the normal heading angle being less than or equal to 1°.

[0013] Optionally, the escape strategy of the intelligent vehicle to reach the target position based on the current position, the transfer grid and the target grid where the target position is located comprises: determining a first original navigation road section based on the normal start position of the transfer grid and the normal start position of the target grid; determine a second escape strategy of the intelligent vehicle from a normal start position of the connection grid to a normal start position of the target grid based on current attribute information of each grid through which the first original navigation path passes in the driving electronic map.

[0014] Optionally, the determining the escape strategy of the intelligent vehicle to reach the target position comprises: when the current grid belongs to the grid through which the original navigation path passes in the driving electronic map, determining a second original navigation path from the current position to a normal start position of the target grid where the target position is located; determine a third escape strategy of the intelligent vehicle to reach the target position on the original navigation path based on the recommended deviation coefficient and the normal driving value of each grid through which the second original navigation path passes in the driving electronic map, wherein the third escape strategy comprises that the intelligent vehicle drives through the corresponding grid based on the recommended deviation coefficient and the normal driving value of each grid through which the original navigation path passes until reaching the target position.

[0015] Optionally, the method further comprises: when it is detected that there is a static obstacle on the connection path, acquiring the current position of the intelligent vehicle and the obstacle position; determining a new connection grid on the original navigation path based on the current position and the obstacle position of the static obstacle, and triggering the step of determining the escape strategy of the intelligent vehicle to reach the target position based on the current position, the connection grid and the target grid where the target position is located.

[0016] Optionally, determining a new connection grid on the original navigation path based on the current position and the obstacle position of the static obstacle comprises: when the distance value between the current position and the obstacle position is greater than or equal to a preset safety distance threshold, acquiring appearance information of the static obstacle collected by the intelligent vehicle and / or appearance information of the static obstacle collected by at least one intelligent vehicle passing by, and generating an obstacle local model of the static obstacle in the driving electronic map based on all the appearance information; determining a detection position of the intelligent vehicle based on the driving electronic map; controlling the intelligent vehicle to move to the detection position to collect new appearance information of the static obstacle; supplementing the obstacle local model based on the new appearance information; determining a new connection grid based on the current position represented by the detection position, the original navigation path and the supplemented obstacle local model.

[0017] Optionally, the method further comprises: determining a nearest grid in the original navigation path based on the detected position of the intelligent vehicle; projecting a straight line connecting the detected position and a normal start position of the nearest grid onto a preset horizontal plane of the driving electronic map, and projecting the local obstacle model onto the preset horizontal plane; when the straight line does not intersect with the projected line of the local obstacle model, determining the nearest grid as the new connection grid.

[0018] Optionally, the method further comprises: when the distance value between the current position and the obstacle position is less than a preset safety distance threshold, prompting a warning.

[0019] Optionally, the method further comprises: if a preset number of grids are continuously driven through, and the actual pose information after driving through each grid is corrected based on the normal end pose information of the respective grid, prompting a warning.

[0020] According to the specific embodiments of the present application, the second aspect, the present application provides a control system, comprising: an intelligent vehicle configured to: send a start application when receiving a start control instruction of an upper computer; and execute a get-out-of-trouble control instruction when receiving the get-out-of-trouble control instruction fed back in response to the start application; an upper computer configured to generate a start control instruction of the intelligent vehicle based on a production task, upload the start application, and download the get-out-of-trouble control instruction; a group control device having a computer program stored thereon and configured to generate a production task based on a production plan and a production progress, and execute the computer program to implement the method as described above.

[0021] The above scheme of the embodiments of the present application has at least the following beneficial effects compared with the prior art: The application provides a method and a control system for escaping from a trouble of an intelligent small vehicle. The application acquires a current position of the intelligent small vehicle in suspension; determines a current grid in which the intelligent small vehicle is located in a driving electronic map and a target position of a target two-dimensional code on an original navigation path based on the current position; and determines an escaping strategy of the intelligent small vehicle to the target position based on a positional relationship between the current grid and the original navigation path in the driving electronic map, in combination with the current position and attribute information of grids on the original navigation path. The driving electronic map is used for compensation to assist the intelligent small vehicle to escape from the trouble, so that the intelligent small vehicle can accurately return to the navigation path from the trouble even in a complex environment, the difficulty of escaping from the trouble is reduced, and the efficiency of escaping from the trouble is improved; unified planning avoids the risk of blind movement of the intelligent small vehicle; manual intervention is reduced, the whole process is automated, and the operation and maintenance efficiency is greatly improved; automatic cognition of the environment and self-correction improve the intelligent level. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 A flow chart of a method for escaping from a trouble of an intelligent small vehicle according to an embodiment of the application is shown; Figure 2 A device schematic block diagram of a control system according to an embodiment of the application is shown. DETAILED DESCRIPTION

[0023] In order to make the objectives, technical solutions and advantages of the application clearer, the application will be further described in detail below with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the application, but not all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the application.

[0024] The terms used in the embodiments of the application are only for the purpose of describing the specific embodiments, and are not intended to limit the application. The singular forms "a", "an" and "the" used in the embodiments of the application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. "Multiple" generally includes at least two.

[0025] It should be understood that the term "and / or" used herein only describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " herein generally represents an "or" relationship between the front and rear associated objects.

[0026] It should be understood that, although the terms first, second, third, etc. can be employed in this application to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of the present application.

[0027] Depending on the context, the word "if" as used herein can be interpreted to mean "when" or "while" or "in response to determining" or "in response to detecting." Similarly, the phrase "if it is determined" or "if [a stated condition or event] is detected" can be interpreted to mean "upon determining" or "in response to determining" or "upon detecting [the stated condition or event]" or "in response to detecting [the stated condition or event]."

[0028] It is also to be noted that the terms "comprising", "including", and any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises a... " does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0029] It is particularly noted that symbols and / or numbers present in the specification, if not marked in the description of the drawings, are not drawing reference numbers.

[0030] The optional embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0031] Embodiment 1 The embodiments provided in the present application are an embodiment of a method for an intelligent small car to escape from a trouble.

[0032] The following will be described in detail Figure 1 The embodiments of the present application will be described in detail.

[0033] In step S101, the current position of the intelligent small car is obtained.

[0034] The intelligent small car can be an automated guided vehicle (English full name: Automated Guided Vehicle, abbreviated as AGV) or IGV.

[0035] The embodiments of the present application provide a method for an intelligent small car to escape from a trouble when staying on a non-matrix two-dimensional code.

[0036] The current position where the intelligent vehicle pauses can be uploaded to the group control device by the sensor (such as laser radar) after the intelligent vehicle detects the current position after receiving the starting instruction and triggering the escape condition, or the current position of the intelligent vehicle can be detected by the positioning equipment installed on site by the group control device after the intelligent vehicle sends an escape application to the group control device after receiving the starting instruction and triggering the escape condition, and the application is not limited to this.

[0037] For example, the escape condition includes that the intelligent vehicle detects that no matrix two-dimensional code is recognized for two consecutive control periods (such as a control period greater than or equal to 0.2s) and the vehicle speed is equal to zero.

[0038] In step S102, the current grid where the intelligent vehicle is located in the driving electronic map and the target position of the target two-dimensional code on the original navigation path are determined based on the current position.

[0039] The original navigation path refers to the navigation path planned in this navigation.

[0040] The target two-dimensional code is a matrix two-dimensional code that the intelligent vehicle has not traveled to in the original navigation path.

[0041] Optionally, the target two-dimensional code is the closest matrix two-dimensional code that the intelligent vehicle has not traveled to in the original navigation path. For example, the intelligent vehicle pauses during the process of driving from the first matrix two-dimensional code to the second matrix two-dimensional code on the original navigation path, and if it stays between the first matrix two-dimensional code and the second matrix two-dimensional code, the second matrix two-dimensional code is the target two-dimensional code.

[0042] The driving electronic map is a grid-based electronic map. The driving electronic map includes an ideal kinematic model, which is a normal driving value obtained when the intelligent vehicle travels on the navigation path in an ideal state.

[0043] The ideal state is a perfect state in which all conditions and results are in the best state. For the embodiments of the present application, the ideal state refers to the state in which the intelligent vehicle normally travels under the condition that the ground is normal. During the multiple normal travels of the intelligent vehicle, multiple historical driving values (such as linear speed, angular speed, and turning radius) of the driving wheels in each grid are recorded, the normal driving values of the corresponding grids are initialized through the multiple historical driving values in each grid, and the ideal kinematic model in the driving electronic map is established based on the normal driving values of each grid. If there is no environmental interference, the intelligent vehicle can normally travel along the navigation path, at this time, the intelligent vehicle does not need the driving electronic map to identify the ground condition, nor does it need to rely on the driving electronic map to assist in driving, and it can travel autonomously. If the environment is abnormal, the intelligent vehicle needs to identify the ground condition by using the driving electronic map, and assist in driving by relying on the driving electronic map to compensate for the driving command and adjust the driving value of the driving wheel, so as to safely and reliably drive through the abnormal area, at the same time, the driving data and environmental data of the safe driving are provided to the driving electronic map, the driving electronic map improves the adaptability to the environment through the self-learning and updating ability, and increases the online updating mechanism of the driving electronic map. The subsequent vehicle can quickly identify the ground abnormality, eliminate abnormal interference, and drive through the interference environment as soon as possible, thereby improving the driving efficiency.

[0044] The driving data related to the abnormal environment is stored in the attribute information of each grid related to the navigation path in the driving electronic map, including the recommended deviation coefficient and the normal driving value. The recommended deviation coefficient is a variable closely related to the driving environment of the grid. The normal driving value is initialized by collecting the historical driving value (such as linear speed value, angular speed value, and torque value) of the intelligent car on the ideal path, and the recommended deviation coefficient is dynamically updated based on the deviation between the actual driving value and the normal driving value. The recommended deviation coefficient can provide compensation driving information for the intelligent car driving on the grid, so that the intelligent car can obtain the compensated current actual driving value through the recommended deviation coefficient, so that the intelligent car can smoothly pass through the grid providing attribute information through the current actual driving value. The driving electronic map creates a digital twin driving model of the intelligent car, which contains the influence information of the physical characteristics of the environment on the vehicle motion, so that the intelligent car has the memory of the complex environment. For example, when it is determined that the current position information of the intelligent car is at the intersection of the first grid and the second grid in the driving electronic map, the current actual deviation coefficient in the first grid and the current recommended deviation coefficient saved in the first grid are obtained; when the absolute value of the difference between the current actual deviation coefficient in the first grid and the current recommended deviation coefficient saved in the first grid is greater than the preset similar coefficient threshold, the normal driving value saved in the second grid is obtained; the product value of the normal driving value saved in the second grid and the current actual deviation coefficient in the first grid is calculated, and the normal driving value saved in the second grid is added to obtain the current actual driving value in the second grid; the intelligent car is driven to drive in the second grid based on the current actual driving value. The recommended deviation coefficient is an empirical coefficient learned based on historical data, which is used to predict the driving compensation required when driving in a specific grid. Its initial value can be calculated by the formula recommended deviation coefficient = (historical actual driving value - normal driving value) / normal driving value, and is updated after the intelligent car successfully passes through the grid each time.

[0045] The driving value is a set of quantitative data of the running state of the driving system (especially the driving wheel) of the intelligent car during driving, which directly reflects how the driving wheel outputs power to realize the driving process, and is the key basis for judging whether the driving wheel slips, the driving force is stable, and the driving system is normal.

[0046] The driving value of the intelligent car mainly includes three types of key data, each of which corresponds to a specific dimension of the driving wheel operation: driving wheel motion state data, driving wheel power output data, and driving system electrical parameter value.

[0047] The driving wheel motion state data directly reflects the physical motion characteristics of the driving wheel, and is the basis for calculating the driving distance and speed of the intelligent car, including: linear speed value, normal angular speed value, and driving distance value.

[0048] The driving wheel power output data reflects the size of the driving wheel output power, and is directly related to whether the driving force of the intelligent vehicle is sufficient to cope with complex environments (such as wet ground and load changes). The core is torque. Torque is the "power intensity" of the driving wheel driving the intelligent vehicle, which is collected through a torque sensor. For example, in the wet ground scenario of the back-to-south season, if the intelligent vehicle needs to maintain the original speed, the driving wheel needs to output more torque to overcome the problem of reduced ground friction; if the torque increases but the speed does not increase synchronously, it means that the driving wheel is slipping (power is not effectively converted into driving speed), which needs to trigger map correction.

[0049] The driving system electrical parameter value is an auxiliary judgment basis for the running state of the driving system, used to troubleshoot whether the driving deviation is caused by electrical failure, including current and / or voltage.

[0050] The mapping relationship between the position and the grid is established in the driving electronic map. The global coordinates (x, y) of the intelligent vehicle are substituted into the coordinate range formula of the map grid to calculate the current grid ID: if the grid size is L (such as 10 cm x 10 cm), the origin coordinates of the driving electronic map are (x0, y0), then the column number col = (x - x0) / L, the row number row = (y - y0) / L, and the grid ID = row-col. For example, x = 125.32 m, x0 = 100 m, and L = 0.1 m, then col = (125.32 - 100) / 0.1) = 253, and row is obtained in the same way.

[0051] In step S103, based on the position relationship between the current grid and the original navigation path in the driving electronic map, the current position, and the attribute information of the grid on the original navigation path, a strategy for the intelligent vehicle to reach the target position is determined.

[0052] The embodiment of the present application assists the intelligent vehicle in escaping from the trap through the grid in the driving electronic map.

[0053] The embodiment of the present application formulates two escape strategies.

[0054] In some specific embodiments, the determination of the escape strategy for the intelligent vehicle to reach the target position comprises: In step S103a-1, when the current grid is not a grid in the original navigation path in the driving electronic map, a connection grid on the original navigation path is determined.

[0055] The connection grid is a grid in the original navigation path, used as a transfer grid from the current position to the original navigation path.

[0056] The current grid is not the grid in the original navigation path in the driving electronic map, which can be understood as that the intelligent vehicle has left the original navigation path. That is, the attribute information of the current grid does not include the recommended deviation coefficient and the normal driving value, that is, no intelligent vehicle has traveled to the current grid in the past. Therefore, the present embodiment proposes a two-step escape strategy through the connection grid.

[0057] Determining the connection grid on the original navigation path and then reaching the target position by using the attribute information of the grid in the original navigation path can reduce the difficulty of escape and improve the efficiency of escape.

[0058] In some embodiments, the determining the connection grid on the original navigation path comprises: Step S103a-1a, determining the grid closest to the current position in the original navigation path as the connection grid, when the recommended deviation coefficient in the original navigation path meets the preset coefficient screening condition.

[0059] The preset coefficient screening condition is used to screen the grid with favorable environmental factors for the intelligent vehicle to continue driving from the grid related to the original navigation path as the connection grid.

[0060] The present embodiment uses the distance and the recommended deviation coefficient as the condition for determining the connection grid, further reducing the difficulty of escape and improving the efficiency of escape.

[0061] In some embodiments, the determining the grid closest to the current position in the original navigation path as the connection grid, when the recommended deviation coefficient in the original navigation path meets the preset coefficient screening condition, comprises: Step S103a-1a-1, determining the closest grid in the original navigation path based on the current position.

[0062] Step S103a-1a-2, determining the original navigation section from the closest grid to the target grid in the original navigation path.

[0063] Step S103a-1a-3, determining a plurality of grids with a preset number of grids before the closest grid in the original navigation section based on the driving direction.

[0064] Step S103a-1a-4, screening at least one candidate grid with the smallest recommended deviation coefficient from the plurality of grids and the closest grid.

[0065] Step S103a-1a-1, step S103a-1a-2, step S103a-1a-3 and step S103a-1a-4 all belong to the preset coefficient screening condition.

[0066] Step S103a-1a-5, determining the candidate grid closest to the current position from the at least one candidate grid as the transfer grid.

[0067] For example, the A0 grid is the closest grid, and the A0 grid is also the last grid in the original navigation section. If the preset number of grids is 5, the A1 grid, the A2 grid, the A3 grid, the A4 grid, and the A5 grid are in turn closest to the A0 grid. The recommended deviation coefficient of the A1 grid is 0.3, the recommended deviation coefficient of the A2 grid is 0.2, the recommended deviation coefficient of the A3 grid is 0.4, the recommended deviation coefficient of the A4 grid is 0.3, and the recommended deviation coefficient of the A5 grid is 0.2. Among them, the recommended deviation coefficients of the A2 grid and the A5 grid are both minimum (i.e. 0.2), but the A2 grid is closest to the current position, and the A2 grid is determined as the transfer grid.

[0068] The embodiment does not simply select the grid closest to the original navigation path as the transfer grid, but selects multiple grids from the closest grid (i.e. the closest grid to the target position) to select the grid with the minimum recommended deviation coefficient and the closest distance as the transfer grid. The transfer grid is close to the target position and not far from the current position, which is conducive to the smart car quickly returning to the original navigation path and reducing the driving distance on the original navigation path. The smaller the recommended deviation coefficient, the lower the driving difficulty in the corresponding grid. Both the distance and the driving difficulty in the transfer grid are taken into account, the difficulty of escaping is reduced, and the efficiency of escaping is improved.

[0069] Step S103a-2, determining an escape strategy of the smart car to reach the target position based on the current position, the transfer grid, and the target grid where the target position is located.

[0070] The escape strategy includes that the smart car drives from the current position to the transfer grid, and then drives from the transfer grid based on the recommended deviation coefficient and the normal driving value of each grid on the original navigation path to pass through the corresponding grid until reaching the target position.

[0071] The embodiment proposes a two-step escape strategy through the transfer grid, i.e. completing the first step of escaping from the current position to the transfer grid, and completing the second step of escaping from the transfer grid to the target position.

[0072] In some embodiments, the determination of the escape strategy of the smart car to reach the target position based on the current position, the transfer grid, and the target grid where the target position is located includes: Step S103a-21, determining a first escape strategy based on the current position, the normal starting position of the transfer grid, and the target position.

[0073] The first escape strategy includes that the intelligent vehicle drives from the current position to the normal start position of the docking grid, and then drives through the corresponding grid based on the recommended deviation coefficient and the normal driving value of each grid passed on the original navigation path from the normal start position of the docking grid.

[0074] The attribute information of the grid in the driving electronic map further includes a normal start position and a normal end position of an ideal path in an ideal kinematic model at a boundary of the corresponding grid.

[0075] The normal start position and the normal end position are start and end positions of the intelligent vehicle passing through the grid when the intelligent vehicle drives on the navigation path in the ideal state in the ideal kinematic model.

[0076] In some embodiments, the first escape strategy is determined based on the current position, the normal start position of the docking grid, and the target position. Step S103a-211, determining a docking section in the first escape strategy based on the current position and the normal start position of the docking grid.

[0077] Since the docking grid is on the original navigation path, the attribute information of the docking grid further includes a normal start position and a normal end position. The end position of the docking section is set at the normal start position of the docking grid, and a section (i.e., the docking section) is established based on the current position and the normal start position of the docking grid, so that the intelligent vehicle can reach the normal start position of the docking grid by using the docking section, and then drive by using the recommended deviation coefficient and the normal driving value in the docking grid, thereby accurately controlling the driving of the intelligent vehicle by using the driving rules of the driving electronic map.

[0078] Step S103a-212, determining normal start pose information and normal end pose information of each key grid passed by the docking section in the driving electronic map, and the normal start pose information of each key grid is consistent with the normal end pose information of the immediately preceding key grid.

[0079] The pose information includes a position and an attitude.

[0080] The key grid refers to a grid passed by the docking section in the driving electronic map.

[0081] The size of the grid can be set according to an empirical value, for example, can be set to 20 cm, 40 cm, etc.

[0082] The docking section is a section selected between the current position and the normal start position of the docking grid according to a path planning algorithm.

[0083] In step S103a-213, the normal driving value and the recommended deviation coefficient in the docking grid are assigned to the normal driving value and the recommended deviation coefficient of each key grid in the first escape strategy.

[0084] The normal driving value includes a normal linear velocity value of each wheel and a normal angular velocity value of the corresponding wheel, and the recommended deviation coefficient includes a recommended deviation coefficient of each wheel.

[0085] Since the docking grid is relatively close to the current position of the intelligent vehicle and the environment is similar, the normal driving value and the recommended deviation coefficient in the docking grid are assigned to each key grid, so that the intelligent vehicle can drive using the driving data of the similar environment, reduce the calculation complexity, and enable the intelligent vehicle to quickly escape from the complex environment.

[0086] The first escape strategy includes that the intelligent vehicle drives through the corresponding grid based on the recommended deviation coefficient and the normal driving value of each key grid passed on the docking section from the current position, and then drives through the corresponding grid based on the recommended deviation coefficient and the normal driving value of each grid passed on the original navigation path from the normal starting position of the docking grid until reaching the target position.

[0087] In some embodiments, the first escape strategy further includes: The pose verification of the intelligent vehicle is triggered once per key grid, and the pose verification of the intelligent vehicle includes: Strategy one, when driving through a key grid, if the actual pose information of the intelligent vehicle meets the preset similar pose condition, the intelligent vehicle enters the next key grid adjacent to the key grid; Strategy two, when driving through a key grid, if the actual pose information of the intelligent vehicle does not meet the preset similar pose condition, the second actual pose information is corrected based on the normal end pose information of the key grid, and then the intelligent vehicle enters the next key grid adjacent to the key grid.

[0088] In this embodiment, strategy one can improve the driving speed of the intelligent vehicle, and strategy two can ensure the driving accuracy of the intelligent vehicle, thereby ensuring that the intelligent vehicle can accurately reach the docking grid.

[0089] In some embodiments, the preset similar pose condition includes: Condition one, the deviation value between the actual position and the normal end position is less than or equal to 3 cm; and, condition two, the deviation value between the actual heading angle and the normal heading angle is less than or equal to 1°.

[0090] The embodiment guarantees the driving speed and driving accuracy of the intelligent vehicle through the first condition and the second condition, so that the first escape strategy is more suitable for the planned path.

[0091] In some embodiments, the escape strategy of the intelligent vehicle to the target position is determined based on the current position, the transfer grid, and a target grid where the target position is located. In step S103a-22-1, a first original navigation path segment is determined based on the normal start position of the transfer grid and the normal start position of the target grid.

[0092] In step S103a-22-2, a second escape strategy of the intelligent vehicle from the normal start position of the transfer grid to the normal start position of the target grid is determined based on the current attribute information of each grid passed by the first original navigation path segment in the driving electronic map.

[0093] Since the influence information of the environmental physical characteristics on the vehicle motion is contained in each grid on the first original navigation path segment, the intelligent vehicle has the memory and predictability of the complex environment, which significantly improves the navigation robustness and reliability in harsh working conditions, speeds up the escape of the intelligent vehicle, and improves the escape efficiency of the intelligent vehicle.

[0094] In some embodiments, the escape strategy of the intelligent vehicle to the target position is determined based on the current position, the transfer grid, and a target grid where the target position is located. In step S103b-1, a second original navigation path segment is determined based on the current position and the normal start position of the target grid when the current grid belongs to the grid passed by the original navigation path in the driving electronic map.

[0095] In step S103b-2, a third escape strategy of the intelligent vehicle to the target position on the original navigation path is determined based on the current attribute information of each grid passed by the second original navigation path segment in the driving electronic map.

[0096] Since the current grid belongs to the grid passed by the original navigation path in the driving electronic map, the influence information of the environmental physical characteristics on the vehicle motion is contained in each grid on the second original navigation path segment in the original navigation path, so that the intelligent vehicle has the memory and predictability of the complex environment, which significantly improves the navigation robustness and reliability in harsh working conditions, speeds up the escape of the intelligent vehicle, and improves the escape efficiency of the intelligent vehicle.

[0097] During the escape of the intelligent vehicle, the deviation between the actual driving data and the pose data can be continuously collected, and the K coefficient of the corresponding grid in the driving map is adjusted online, so that the map becomes more and more accurate.

[0098] In some embodiments, the method further comprises: Step S111, when detecting that there is a static obstacle on the transfer section, acquiring the current position of the intelligent car and the obstacle position.

[0099] Step S112, determining a new transfer grid on the original navigation path based on the current position and the obstacle position of the static obstacle, and triggering the step of executing the escape strategy of the intelligent car reaching the target position based on the current position, the transfer grid and the target grid where the target position is located.

[0100] In the present embodiment, the step of triggering the execution of the escape strategy of the intelligent car reaching the target position based on the current position, the transfer grid and the target grid where the target position is located, is actually triggering the execution of step S103a-2, which solves the problem of the intelligent car bypassing the static obstacle to reach the target position on the transfer section through step S103a-2 and subsequent steps.

[0101] In some embodiments, determining a new transfer grid on the original navigation path based on the current position and the obstacle position of the static obstacle comprises: Step S112-1a, when the distance value between the current position and the obstacle position is greater than or equal to a preset safety distance threshold, acquiring appearance information of the static obstacle collected by the intelligent car and / or appearance information of the static obstacle collected by at least one intelligent car passing by, and generating an obstacle local model of the static obstacle in the driving electronic map based on all the appearance information.

[0102] For example, the preset safety distance threshold is 1 m.

[0103] The appearance information includes depth information. In the field of computer vision, the depth information refers to the information of the distance of an object in a three-dimensional space from an observer. In computer vision, the position relationship of an object in a three-dimensional space can be accurately understood through depth information, and an appearance model of the object in a three-dimensional space can be fitted through such position relationship. In the present embodiment, the depth information can be acquired through a depth camera, a binocular camera or a laser radar (i.e. LiDAR) arranged at the front part of the intelligent car.

[0104] When the intelligent vehicle recognizes a static obstacle (i.e., the obstacle does not move for 3 seconds) on the navigation path, and the static obstacle exceeds the preset safety ignoring threshold of the system (such as a small object with a height value < 5 cm and a width value < 10 cm), the detection process of the static obstacle is triggered immediately. At this time, the intelligent vehicle will brake urgently at a preset distance from the static obstacle and start modeling. The appearance information of the static obstacle is used to generate a local obstacle model of the static obstacle in the three-dimensional electronic map. Since the intelligent vehicle can only collect the appearance information of the static obstacle at the emergency braking point, the local appearance information of the static obstacle is obtained.

[0105] Step S112-2, determining the detection position of the intelligent vehicle based on the driving electronic map.

[0106] The detection position can be selected near the intelligent vehicle or near the static obstacle.

[0107] In the embodiments of the present application, only partial information of the static obstacle can be obtained at the emergency braking point. If the static obstacle is bypassed, the intelligent vehicle needs to determine multiple detection positions according to the detected environment, and continuously improve the local obstacle model through the appearance information collected at the multiple detection positions, so as to determine the bypass path through the local obstacle model. In this process, the intelligent vehicle needs to continuously determine the detection position according to the current local obstacle model in the exploration process, so as to collect the appearance information of the unknown part of the static obstacle through the detection position.

[0108] Step S112-3, controlling the intelligent vehicle to move to the detection position to collect new appearance information of the static obstacle.

[0109] When the detection position is determined, the intelligent vehicle is instructed to reach the detection position to collect the appearance information of the static obstacle, improve the local obstacle model, and improve the continuity of the local obstacle model.

[0110] Step S112-4, supplementing the local obstacle model based on the new appearance information.

[0111] Step S112-5, determining a new connection grid based on the current position represented by the detection position, the original navigation path, and the supplemented local obstacle model.

[0112] When the appearance information of the static obstacle is collected by the intelligent vehicle, the appearance information of the static obstacle is collected by determining the detection position, the local obstacle model is constantly supplemented and improved, and finally the new connection grid on the original navigation path is determined by the supplemented local obstacle model. For example, based on the current position (i.e. the detection position) of the intelligent vehicle, the nearest grid in the original navigation path is determined, a straight line connecting the current position and the normal starting position of the nearest grid is projected onto a preset horizontal plane driving the electronic map, and the local obstacle model is also projected onto the preset horizontal plane. When the straight line does not intersect with the projection line of the local obstacle model, the nearest grid is determined as the new connection grid.

[0113] In the process of the intelligent vehicle escaping from the trouble, the connection grid on the original navigation path is re-determined by the cooperative work of the intelligent vehicle and the group control device, so as to realize the re-escape from the trouble.

[0114] In some embodiments, the method further comprises: Step S112-1b, when the distance value between the current position and the obstacle position is less than the preset safety distance threshold, prompting a warning.

[0115] In the present embodiment, if the distance between the current position of the intelligent vehicle and the obstacle position is too small, which may affect the normal movement of the intelligent vehicle to the detection position, the alarm information is prompted by sound, light, electricity and image, so as to correct the unsafe factors by manual intervention.

[0116] In some embodiments, the method further comprises: Step S121, if a preset number of grids are continuously driven through, and the actual pose information after driving through each grid is corrected based on the normal end pose information of the respective grid, a warning is prompted.

[0117] In the present embodiment, if the intelligent vehicle driving through the grid is corrected for multiple times (such as a preset number of 3) as needed, which indicates that the current correction is failed, the alarm information is prompted by sound, light, electricity and image, so as to correct by manual intervention. For example, it is raining and the ground is full of water.

[0118] The embodiment of the application obtains a current position of the intelligent trolley; determines a current grid in which the intelligent trolley is located in the driving electronic map and a target position of a target two-dimensional code on an original navigation path based on the current position; and determines a strategy for the intelligent trolley to reach the target position based on a positional relationship between the current grid and the original navigation path in the driving electronic map, in combination with the current position and attribute information of the grid on the original navigation path. The driving electronic map is used for compensation to assist the intelligent trolley in escaping from the predicament, so that the intelligent trolley can still accurately return to the navigation path from the predicament even in a complex environment, the difficulty of escaping from the predicament is reduced, and the efficiency of escaping from the predicament is improved; unified planning avoids the risk of blind movement of the intelligent trolley; manual intervention is reduced, full-process automation is realized, and the operation and maintenance efficiency is greatly improved; automatic cognition of the environment and self-correction improve the intelligent level.

[0119] The application also provides a control system embodiment for implementing the method steps of the above embodiments, based on the same explanation of the meanings of the names and the same technical effects as the above embodiments, which will not be described here.

[0120] Embodiment 2 As shown in Figure 2 The embodiment of the application provides a control system, which comprises: an intelligent trolley 21 configured to send a start application when receiving a start control instruction of an upper computer, and execute a predicament-escaping control instruction when receiving the predicament-escaping control instruction fed back in response to the start application; an upper computer 22 configured to generate a start control instruction of the intelligent trolley based on a production task, upload the start application, and download the predicament-escaping control instruction; a group control device 23 having a computer program stored thereon and configured to generate a production task based on a production plan and a production progress, and execute the computer program to implement the method described above.

[0121] In the embodiment of the application, the production plan is made according to an MES system and a production rolling plan and a bill of materials, and a production task of each process island is generated based on the production plan and a production progress, which includes a transportation task of the intelligent trolley. The upper computer 12 is integrated with production line equipment (including the intelligent trolley) to obtain equipment states (running, shutdown, and fault), and generate a control instruction (including a predicament-escaping control instruction of the intelligent trolley) based on the production task. The intelligent trolley executes the production task according to the control instruction, and escapes from the predicament according to the predicament-escaping control instruction issued during the execution process.

[0122] The error control instruction indicates that the intelligent trolley 23 avoids the influence of an adverse factor on the driving path on driving.

[0123] It should be noted that the various embodiments described in the specification are intended to be illustrative only and are not in any way limiting. Although the present application has been described in considerable detail, various modifications and changes can be made to the present application by one skilled in the art and it is intended that the application not be limited to any one embodiment, but encompass numerous alternatives, modifications and equivalents. Therefore, the scope of the application should be determined by the broadest permissible interpretation of the claims so as to encompass all such alternatives, modifications and equivalents.

[0124] The above embodiments are only used to illustrate the technical solutions of the present application, but not limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalent replacements; and the modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for an intelligent vehicle to escape from difficult situations, characterized in that, include: Get the current position where the smart car is paused; Based on the current location, determine the target location of the target QR code on the current grid and the original navigation path of the intelligent vehicle in the driving electronic map, wherein the target QR code is a matrix QR code that the intelligent vehicle has not yet reached in the original navigation path; Based on the positional relationship between the current grid and the original navigation path in the driving electronic map, and combined with the current location and the attribute information of the grid on the original navigation path, the escape strategy for the intelligent vehicle to reach the target location is determined.

2. The method according to claim 1, characterized in that, The escape strategy for determining when the intelligent vehicle reaches the target location includes: When the current grid is not a grid in the original navigation path of the driving electronic map, a connecting grid is determined on the original navigation path; Based on the current location, the connecting grid, and the target grid where the target location is located, an escape strategy for the intelligent vehicle to reach the target location is determined. The escape strategy includes: the intelligent vehicle travels from the current location to the connecting grid, and then, starting from the connecting grid, drives through the corresponding grid based on the recommended deviation coefficient and normal driving value of each grid passed on the original navigation path, until it reaches the target location.

3. The method according to claim 2, characterized in that, Determining the connecting grid on the original navigation path includes: The grid in the original navigation path whose recommended deviation coefficient meets the preset coefficient filtering conditions and is closest to the current position is identified as the connecting grid.

4. The method according to claim 3, characterized in that, The step of determining the grid in the original navigation path whose recommended deviation coefficient meets the preset coefficient filtering conditions and is closest to the current position as the connecting grid includes: Determine the nearest grid in the original navigation path based on the current location; Determine the original navigation segment from the nearest grid to the target grid from the original navigation path; In the original navigation segment, multiple grids with a preset number of grids are determined based on the driving direction before the nearest grid; Select at least one candidate grid with the smallest recommendation deviation coefficient from the plurality of grids and the nearest grid; The candidate grid closest to the current position is determined from the at least one candidate grid as the connecting grid.

5. The method according to claim 2, characterized in that, The strategy for determining the escape route of the intelligent vehicle to reach the target location based on the current location, the connecting grid, and the target grid where the target location is located includes: Based on the current location, the normal starting point of the connecting grid, and the target location, a first escape strategy is determined, wherein the first escape strategy includes: the intelligent vehicle moves from the current location to the normal starting point of the connecting grid, and then, starting from the normal starting point of the connecting grid, drives through the corresponding grid based on the recommended deviation coefficient and normal driving value of each grid passed on the original navigation path, until it reaches the target location.

6. The method according to claim 5, characterized in that, The determination of the first escape strategy based on the current position, the normal starting position of the connection grid, and the target position includes: The connecting road segment in the first escape strategy is determined based on the current location and the normal starting point location of the connecting grid; Determine the normal starting point pose information and normal ending point pose information of each key grid that the connecting road segment passes through in the driving electronic map, and the normal starting point pose information of each key grid is consistent with the normal ending point pose information of the preceding adjacent key grid. The normal driving value and recommended deviation coefficient in the connecting grid are used as the normal driving value and recommended deviation coefficient of each key grid in the first escape strategy. The normal driving value includes the normal linear velocity value and the normal angular velocity value of each wheel, and the recommended deviation coefficient includes the recommended deviation coefficient of each wheel. The first escape strategy includes: the intelligent vehicle starts from the current position and drives through the corresponding grid based on the recommended deviation coefficient and normal driving value of each key grid passed on the connecting road segment until it reaches the normal starting position of the connecting grid. Then, starting from the normal starting position of the connecting grid, it drives through the corresponding grid based on the recommended deviation coefficient and normal driving value of each grid passed on the original navigation path until it reaches the target position.

7. The method according to claim 6, characterized in that, The first escape strategy also includes: Each time the vehicle passes through a key grid, a pose verification of the intelligent vehicle is triggered. This pose verification includes: When the vehicle passes through a key grid, if the actual pose information of the intelligent vehicle meets the preset similar pose conditions, it will enter the next key grid adjacent to the key grid. When passing through a key grid, if the actual pose information of the intelligent vehicle does not meet the preset similar pose conditions, the actual pose information is corrected based on the normal endpoint pose information of the key grid before entering the next key grid adjacent to the key grid.

8. The method according to claim 7, characterized in that, The preset similar pose conditions include: The deviation between the actual position and the normal endpoint position is less than or equal to 3cm; And, the deviation between the actual heading angle and the normal heading angle is less than or equal to 1°.

9. The method according to claim 2, characterized in that, The strategy for determining the escape route of the intelligent vehicle to reach the target location based on the current location, the connecting grid, and the target grid where the target location is located includes: The first original navigation segment is determined based on the normal starting position of the connecting grid and the normal starting position of the target grid; Based on the current attribute information of each grid traversed by the first original navigation route segment in the driving electronic map, a second escape strategy is determined for the intelligent vehicle to reach the normal starting position of the target grid from the normal starting position of the connecting grid.

10. The method according to claim 1, characterized in that, The escape strategy for determining when the intelligent vehicle reaches the target location includes: When the current grid belongs to the grid traversed by the original navigation path in the driving electronic map, a second original navigation segment is determined from the current position to the normal starting position of the target grid where the target position is located; Based on the recommended deviation coefficient and normal driving value of each grid passed by the second original navigation segment in the driving electronic map, a third escape strategy is determined for the intelligent vehicle to reach the target location on the original navigation path. The third escape strategy includes: the intelligent vehicle drives through the corresponding grid based on the recommended deviation coefficient and normal driving value of each grid passed by the original navigation path until it reaches the target location.

11. The method according to claim 6, characterized in that, The method further includes: When a static obstacle is detected on the connecting road section, the current position of the intelligent vehicle and the position of the obstacle are obtained; On the original navigation path, a new connecting grid is determined based on the current position and the obstacle position of the static obstacle, and the step of determining the escape strategy of the intelligent vehicle to reach the target position based on the current position, the connecting grid and the target grid where the target position is located is triggered.

12. The method according to claim 11, characterized in that, Determining a new connection grid on the original navigation path based on the current position and the obstacle positions of the static obstacles includes: When the distance between the current position and the obstacle position is greater than or equal to a preset safe distance threshold, the appearance information of the static obstacle collected by the intelligent vehicle and / or the appearance information of the static obstacle collected by at least one intelligent vehicle passing by are obtained, and a local obstacle model of the static obstacle is generated in the driving electronic map based on all appearance information. The detection location of the intelligent vehicle is determined based on the driving electronic map; The intelligent vehicle is controlled to move to the detection position to collect new appearance information of the static obstacle; The local model of the obstacle is supplemented based on the new appearance information; A new connection grid is determined based on the current position represented by the detected position, the original navigation path, and the supplemented local obstacle model.

13. The method according to claim 12, characterized in that, The process of determining a new connection mesh based on the current position represented by the detected position, the original navigation path, and the supplemented local obstacle model includes: The nearest grid in the original navigation path is determined based on the detected position of the intelligent vehicle; The line connecting the detection location to the normal starting point of the nearest grid is projected onto a preset horizontal plane of the driving electronic map, and the local model of the obstacle is also projected onto the preset horizontal plane. When the straight line does not intersect with the projection line of the local model of the obstacle, the nearest grid is determined to be the new connecting grid.

Images