Flat car unmanned navigation control system based on environmental characteristics

Abstract

The invention discloses a flat car unmanned navigation control system based on environmental characteristics, which belongs to the technical field of unmanned logistics and comprises a safety module and a server. The navigation module is used for carrying out driving navigation on the tablet unmanned vehicle; the safety module is used for performing safety early warning control in the running process of the tablet unmanned vehicle, establishing an early warning area, a deceleration area and a parking area of the tablet unmanned vehicle, acquiring a navigation route map of the tablet unmanned vehicle, positioning the position of the tablet unmanned vehicle in real time, and updating the early warning area, the deceleration area and the parking area of the tablet unmanned vehicle in the navigation route map in real time; real-time obstacle information acquisition is carried out on a navigation route through an acquisition device, when no obstacle is acquired, operation is not carried out, and when the obstacle is acquired, safety control is carried out according to the position of the obstacle; through mutual cooperation of the safety module and the navigation module, safety early warning of the tablet unmanned vehicle in the navigation driving process is realized, and obstacles can be found in time when encountering the obstacles.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned logistics, and specifically relates to an unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics. Background Art

[0002] With the rapid development of intelligent warehousing and intelligent logistics, more and more unmanned vehicles are applied to warehousing logistics. The flatbed unmanned vehicle is one of them. However, unmanned vehicles are extremely likely to encounter obstacles during operation. If not handled properly, safety problems are likely to occur. Therefore, in order to solve this problem, the present invention provides an unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics. Summary of the Invention

[0003] In order to solve the problems existing in the above solutions, the present invention provides an unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics.

[0004] The object of the present invention can be achieved by the following technical solutions:

[0005] An unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics includes a safety module and a server;

[0006] The navigation module is used for driving navigation of the flatbed unmanned vehicle;

[0007] The safety module is used for safety warning control during the driving process of the flatbed unmanned vehicle. The specific method includes:

[0008] Establish a warning area, a deceleration area, and a parking area for the flatbed unmanned vehicle, obtain the navigation route map of the flatbed unmanned vehicle, real-time locate the position of the flatbed unmanned vehicle, and real-time update the warning area, deceleration area, and parking area of the flatbed unmanned vehicle in the navigation route map;

[0009] Collect real-time obstacle information on the navigation route through a collection device. When no obstacle is collected, no operation is performed. When an obstacle is collected, safety control is performed according to the position of the obstacle.

[0010] Further, the method for establishing a warning area, a deceleration area, and a parking area for the flatbed unmanned vehicle includes:

[0011] Based on big data, obtain the historical operation data of the unmanned vehicle, perform data statistics on the obtained historical operation data of the unmanned vehicle to obtain a corresponding operation statistical table, perform data conversion according to the obtained operation statistical table to obtain a classification data set. The classification data set includes a warning data set, a deceleration data set, and a parking data set; analyze the obtained classification data set to obtain corresponding classification calculation formulas, real-time obtain the speed of the flatbed unmanned vehicle, and dynamically generate corresponding warning areas, deceleration areas, and parking areas through the classification calculation formulas according to the obtained speed of the flatbed unmanned vehicle.

[0012] Furthermore, the method for analyzing the obtained classified data set includes:

[0013] Set the standard contours of the warning area, deceleration area, and parking area respectively, marked as the warning standard contour, deceleration standard contour, and parking standard contour. Input the warning data set, deceleration data set, and parking data set into the corresponding coordinate systems respectively, identify the distribution of each classified data set in the corresponding coordinate systems, conduct corresponding analyses, obtain the corresponding classified initial values and classification coefficients, marked as CFz and FLz, where z includes a, b, c, and a, b, c represent the warning area, deceleration area, and parking area respectively. Then the classification calculation formula is LDv = CFz + FLz × V, where V is the vehicle speed of the flat unmanned vehicle, and LDv is the size value of the corresponding classification.

[0014] Furthermore, the method for safety control according to the position of the obstacle includes:

[0015] When the obstacle is outside the warning area, no operation is performed. When the obstacle is within the warning area, an alarm warning is issued to prompt the corresponding obstacle to avoid. When the obstacle is within the deceleration area, control the flat unmanned vehicle to decelerate and issue a continuous alarm warning. When the obstacle is within the parking area, control the flat unmanned vehicle to stop, issue a continuous alarm warning, and send the obstacle information to the corresponding management personnel.

[0016] Furthermore, it further includes a dynamic adjustment module, and the dynamic adjustment module is used to adjust the warning area, deceleration area, and parking area generated by the safety module.

[0017] Furthermore, the working method of the dynamic adjustment module includes:

[0018] Obtain the working area map of the flat unmanned vehicle, obtain the obstacle encounter record of the flat unmanned vehicle in real time, convert the obtained obstacle encounter record into corresponding obstacle encounter points, and mark them at the corresponding positions on the working area map. Conduct area division according to the distribution of the obstacle encounter points to obtain several correction areas; set corresponding classification correction coefficients for each correction area;

[0019] Identify the position of the flat unmanned vehicle in real time, match the corresponding classification correction coefficient according to the identified position of the flat unmanned vehicle, obtain the size value LDv of the current flat unmanned vehicle, and according to the formula LDXv = LDv × βv, where βv is the corresponding classification correction coefficient, adjust the warning area, deceleration area, and parking area according to the calculated corrected size value.

[0020] Furthermore, the specific method for area division according to the distribution of the obstacle encounter points includes:

[0021] Identify the passage areas in the working area map; divide the identified passage areas into several unit areas, identify the number of obstacle points in each unit area, classify the unit areas according to the identified number of obstacle points, and merge them according to the unit area levels to obtain several corrected areas.

[0022] Further, the method for classifying unit areas according to the identified number of obstacle points includes:

[0023] Set several quantity intervals, set corresponding area levels for each quantity interval, and establish a corresponding level matching table according to the set quantity intervals and corresponding area levels;

[0024] Obtain the obstacle point data of each unit area, match the corresponding area level from the level matching table according to the obtained obstacle point data, and complete the classification of the unit area.

[0025] Further, the method for setting corresponding classification correction coefficients for each corrected area includes:

[0026] Statistically analyze the quantity proportion of the obstacle types represented by the obstacle points in each corrected area, establish a corresponding coefficient analysis model, and analyze the quantity proportion of each obstacle type in the corrected area and the corrected area level through the established coefficient analysis model to obtain the corresponding classification correction coefficients.

[0027] Compared with the prior art, the beneficial effects of the present invention are:

[0028] Through the mutual cooperation between the safety module and the navigation module, the safety warning of the flat unmanned vehicle during the navigation driving process is realized, so that obstacles can be discovered in time and corresponding emergency treatments can be made; and then through the mutual cooperation with the dynamic adjustment module, the operation data of the flat unmanned vehicle are dynamically analyzed according to the actual situation, and dynamic adjustment is made according to the analysis results, which is more adaptable to each enterprise and improves the corresponding work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0030] Figure 1 It is a principle block diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative work shall fall within the protection scope of the present invention.

[0032] As Figure 1 shown, the driverless navigation control system of the flatbed vehicle based on environmental characteristics includes a safety module, a navigation module, and a server;

[0033] The navigation module is used to perform the driving navigation of the flatbed unmanned vehicle and uses existing path selection algorithms for navigation.

[0034] The safety module is used to perform safety warning control during the driving process of the flatbed unmanned vehicle, that is, mainly used for real-time obstacle scanning warning and corresponding safety control during the driving process of the flatbed unmanned vehicle on the path to avoid the occurrence of safety accidents; the specific methods include:

[0035] Establish the warning area, deceleration area, and parking area of the flatbed unmanned vehicle, obtain the navigation route map of the flatbed unmanned vehicle, real-time locate the position of the flatbed unmanned vehicle, and update the warning area, deceleration area, and parking area of the flatbed unmanned vehicle in the navigation route map in real-time;

[0036] Collect real-time obstacle information on the navigation route through a collection device, such as collecting obstacles by using a laser obstacle sensor and a radar scanner. When no obstacles are collected, no operation is performed. When obstacles are collected, safety control is performed according to the position of the obstacles.

[0037] The methods for establishing the warning area, deceleration area, and parking area of the flatbed unmanned vehicle include:

[0038] Obtain a large amount of historical operation data of the unmanned vehicle based on big data, perform data statistics on the obtained historical operation data of the unmanned vehicle to obtain a corresponding operation statistical table, perform data conversion according to the obtained operation statistical table to obtain a classification data set, and the classification data set includes a warning data set, a deceleration data set, and a parking data set; analyze the obtained classification data set to obtain corresponding classification calculation formulas, real-time obtain the speed of the flatbed unmanned vehicle, and dynamically generate the corresponding warning area, deceleration area, and parking area according to the obtained speed of the flatbed unmanned vehicle through the classification calculation formulas.

[0039] The historical operation data of the driverless vehicle can be the historical operation data of the flatbed driverless vehicle, or the historical operation data of other driverless vehicles in the same or similar working environments. When the direct historical operation data of the flatbed driverless vehicle is insufficient, other historical operation data can be used for analysis. The historical operation data of the driverless vehicle includes obstacle types, response data, whether a safety accident has occurred, etc. The response data refers to response measures such as warning, deceleration, and stopping.

[0040] The operation statistical table sets the corresponding statistical template by means of discussion by an expert group, and uses existing statistical methods to conduct data statistics according to the statistical template, such as the obstacle types, the share, the response time corresponding to the response measures, the vehicle speed, whether the response purpose is achieved, etc.

[0041] Data conversion is performed according to the obtained operation statistical table. The corresponding statistical data is classified and split according to the warning, deceleration, and stopping in the response measures to obtain the statistical data of the corresponding classification. The obtained classified statistical data is converted according to a preset conversion scheme to obtain the corresponding classified data set.

[0042] Converting the obtained classified statistical data according to a preset conversion scheme means that, according to the possible data, a corresponding numerical conversion relationship is set manually, a corresponding conversion table is established, and the conversion is completed after corresponding matching to form a corresponding coordinate form, and the corresponding classified data set is obtained after summarization.

[0043] In one embodiment, the method for analyzing the obtained classified data set includes:

[0044] The standard outlines of the warning area, deceleration area, and parking area are set respectively, and are marked as the warning standard outline, deceleration standard outline, and parking standard outline. Generally, they are set as rectangles, which include the synchronous proportional relationship of each side of the corresponding rectangle, that is, when the size in one direction is confirmed, the corresponding area shape is automatically generated. It can also be an arc, etc. Specifically, it is set by means of discussion by an expert group; the warning data set, deceleration data set, and parking data set are respectively input into the corresponding coordinate system, the distribution of each classified data set in the corresponding coordinate system is identified, and corresponding analysis is performed to obtain the corresponding classified initial value and classification coefficient, marked as CFz and FLz, where z includes a, b, c, and a, b, c respectively represent the warning area, deceleration area, and parking area. Then the classification calculation formula is LDv = CFz + FLz × V, where V is the vehicle speed of the flatbed driverless vehicle, and LDv is the size value of the corresponding classification.

[0045] Identify the distribution of each classification dataset in the corresponding coordinate coefficients, perform corresponding analyses, obtain the corresponding classification initial values and classification coefficients, and set the initial values for ensuring safety response and the classification coefficients related to vehicle speed according to the corresponding coordinates through manual analysis. These are used to ensure that no safety accidents occur under normal circumstances through the classification calculation formula in all obstacle situations, which is a unified standard.

[0046] In another embodiment, directly set the corresponding classification calculation formula through existing analysis methods to determine the corresponding warning area, deceleration area, and parking area, that is, determine the corresponding warning area, deceleration area, and parking area through existing methods.

[0047] The method for safety control according to the position of the obstacle includes:

[0048] When the obstacle is outside the warning area, no operation is performed. When the obstacle is within the warning area, an alarm warning is issued to prompt the corresponding obstacle to avoid. When the obstacle is within the deceleration area, control the flat unmanned vehicle to decelerate and continuously issue an alarm warning. When the obstacle is within the parking area, control the flat unmanned vehicle to stop, continuously issue an alarm warning, and send the obstacle information to the corresponding management personnel, who will make corresponding coordination according to the actual situation.

[0049] In one embodiment, since the warning area, deceleration area, and parking area set through the safety module are universal, that is, they are used to adapt to most application scenarios. However, in the actual application process, due to the uniqueness and diversity of the enterprise work environment, the warning area, deceleration area, and parking area still have a large adjustment range to make them more suitable for the use of the corresponding enterprise and improve the corresponding transportation efficiency. Therefore, a dynamic adjustment module is set up to adjust the warning area, deceleration area, and parking area generated by the safety module. The specific method includes:

[0050] Obtain the working area map of the flat unmanned vehicle, which includes the channel information that the flat unmanned vehicle can pass through; obtain the obstacle encounter record of the flat unmanned vehicle in real time, and the obstacle encounter record includes time, location, and obstacle type; convert the obtained obstacle encounter record into corresponding obstacle encounter points and mark them at the corresponding positions on the working area map, and divide the area according to the distribution of the obstacle encounter points to obtain several correction areas; set corresponding classification correction coefficients for each correction area;

[0051] Identify the position of the flat unmanned vehicle in real time, match the corresponding classification correction coefficient according to the identified position of the flat unmanned vehicle, obtain the size value LDv of the current flat unmanned vehicle, and according to the formula LDXv = LDv × βv, where βv is the corresponding classification correction coefficient, adjust the warning area, deceleration area, and parking area according to the calculated corrected size value.

[0052] Convert the obtained obstacle encounter records into corresponding obstacle encounter points, that is, convert the obstacle encounter records including time, location, and obstacle type into corresponding coordinate points. For corresponding conversion, the conversion method of the classification dataset can be referred to for corresponding conversion.

[0053] Conduct regional division according to the distribution of obstacle encounter points, that is, divide the passage area into several small areas for setting the same adjustment coefficient within the same area in the follow-up; the specific method includes:

[0054] Identify the passage area in the working area map, which refers to the passage area of the flat unmanned vehicle; divide the identified passage area into several unit areas, identify the number of obstacle encounter points in each unit area, conduct grading of the unit areas according to the identified number of obstacle encounter points, and conduct merging according to the unit area grades to obtain several corrected areas.

[0055] Divide the identified passage area into several unit areas according to the number of lanes in the passage area. First, identify and divide the corresponding lane lines, set the width of the unit area manually, and the length is the distance between two lane lines. Divide the passage area according to the definition of the unit area.

[0056] The method for grading unit areas according to the identified number of obstacle encounter points includes:

[0057] Manually set several quantity intervals, set corresponding area grades for each quantity interval, all of which are set through the discussion of the expert group, and establish a corresponding grade matching table according to the set quantity intervals and corresponding area grades;

[0058] Obtain the obstacle encounter point data of each unit area, match the corresponding area grade from the grade matching table according to the obtained obstacle encounter point data, and complete the grading of the unit area.

[0059] Conduct merging according to the unit area grades, that is, merge the unit areas of the same grade as much as possible and conduct merging in combination with the positions of the corresponding unit areas. Specifically, existing area merging methods can be used for area merging.

[0060] The method for setting corresponding classification correction coefficients for each corrected area includes:

[0061] Statistically analyze the quantity ratio of the obstacle types represented by the obstacle encounter points in each corrected area, establish a corresponding coefficient analysis model, analyze the quantity ratio of each obstacle type in the corrected area and the corrected area grade through the established coefficient analysis model, and obtain the corresponding classification correction coefficients. The corrected area grade is determined according to the corresponding unit area grade.

[0062] The coefficient analysis model is established based on the CNN network and the DNN network, trained by manually establishing a corresponding training set, and analyzed by the coefficient analysis model after successful training.

[0063] The above formulas are all calculated by removing the dimension and taking their numerical values. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained by simulating a large amount of data.

[0064] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. The unmanned driving navigation control system of a flatbed vehicle based on environmental characteristics is characterized by: Includes security module and server; The navigation module is used for driving navigation of the flatbed unmanned vehicle; The safety module is used to perform safety warning control during the driving process of the flatbed unmanned vehicle, and the specific method includes: Establish warning zones, deceleration zones, and parking zones for flatbed unmanned vehicles, obtain navigation route maps for flatbed unmanned vehicles, locate the position of flatbed unmanned vehicles in real time, and update warning zones, deceleration zones, and parking zones for flatbed unmanned vehicles in real time in the navigation route maps; Real-time obstacle information on the navigation route is collected through the collection device. When no obstacle is collected, no operation is performed. When an obstacle is collected, safety control is performed according to the location of the obstacle.

2. The unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics according to claim 1 is characterized in that: Methods for establishing warning zones, deceleration zones, and parking zones for flatbed unmanned vehicles include: Based on big data, historical operation data of the unmanned vehicle is obtained, and data statistics are performed on the obtained historical operation data of the unmanned vehicle to obtain a corresponding operation statistics table. Data conversion is performed according to the obtained operation statistics table to obtain a classified data set, and the classified data set includes a warning data set, a deceleration data set, and a parking data set; the obtained classified data set is analyzed to obtain a corresponding classification calculation formula, and the speed of the flatbed unmanned vehicle is obtained in real time. According to the obtained speed of the flatbed unmanned vehicle, the corresponding warning area, deceleration area, and parking area are dynamically generated through the classification calculation formula.

3. The unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics according to claim 2 is characterized in that: Methods for analyzing the obtained classification dataset include: The standard contours of the warning area, deceleration area and parking area are set respectively, and marked as the warning standard contour, deceleration standard contour and parking standard contour respectively. The warning data set, deceleration data set and parking data set are input into the corresponding coordinate system respectively, and the distribution of each classification data set in the corresponding coordinate coefficient is identified. The corresponding analysis is performed to obtain the corresponding classification initial value and classification coefficient, which are marked as CFz and FLz, where z includes a, b, c, and a, b, c represent the warning area, deceleration area and parking area respectively. The classification calculation formula is LDv=CFz+FLz×V, where V is the speed of the flatbed unmanned vehicle, and LDv is the size value of the corresponding classification.

4. The unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics according to claim 3 is characterized in that: Methods for safety control based on the location of obstacles include: When the obstacle is outside the warning area, no operation is performed. When the obstacle is within the warning area, an alarm warning is issued to prompt the corresponding obstacle to be avoided. When the obstacle is within the deceleration zone, the flatbed unmanned vehicle is controlled to slow down and a continuous alarm warning is issued. When the obstacle is within the parking area, the flatbed unmanned vehicle is controlled to park, a continuous alarm warning is issued, and the obstacle information is sent to the corresponding management personnel.

5. The unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics according to claim 3 is characterized in that: It also includes a dynamic adjustment module, which is used to adjust the warning area, deceleration area and parking area generated by the safety module.

6. The unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics according to claim 5 is characterized in that: The working methods of the dynamic adjustment module include: Obtain a working area map of the flatbed unmanned vehicle, obtain the obstacle records of the flatbed unmanned vehicle encountering obstacles in real time, convert the obtained obstacle records into corresponding obstacle points, and mark them at corresponding positions in the working area map, divide the areas according to the distribution of the obstacle points, and obtain several correction areas; set a corresponding classification correction coefficient for each correction area; Identify the position of the flatbed unmanned vehicle in real time, match the corresponding classification correction coefficient according to the identified position of the flatbed unmanned vehicle, obtain the current size value LDv of the flatbed unmanned vehicle, and adjust the warning area, deceleration area and parking area according to the calculated corrected size value according to the formula LDXv=LDv×βv, where βv is the corresponding classification correction coefficient.

7. The unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics according to claim 6 is characterized in that: The area is divided according to the distribution of obstacle points. The specific methods include: Identify the passage area in the work area map; divide the identified passage area into a number of unit areas, identify the number of obstacle points in each unit area, grade the unit area according to the number of identified obstacle points, merge the unit areas according to their levels, and obtain a number of correction areas.

8. The unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics according to claim 7 is characterized in that: Methods for grading unit areas based on the number of identified obstacles include: Set a number of quantity intervals, set a corresponding regional level for each quantity interval, and establish a corresponding level matching table according to the set quantity intervals and the corresponding regional levels; Obtain the obstacle point data of each unit area, match the corresponding area level from the level matching table according to the obtained obstacle point data, and complete the grading of the unit area.

9. The unmanned driving navigation control system for a flatbed vehicle based on environmental characteristics according to claim 8 is characterized in that: The method of setting the corresponding classification correction coefficient for each correction area includes: The number ratio of obstacle types represented by the obstacle points in each correction area is counted, and a corresponding coefficient analysis model is established. The number ratio of each obstacle type in the correction area and the correction area level are analyzed through the established coefficient analysis model to obtain the corresponding classification correction coefficient.

Images