A fusion navigation method and device for an unmanned vehicle

Abstract

The present invention discloses a fusion navigation method and device for an autonomous vehicle. The fusion navigation method includes: driving the autonomous vehicle to travel along the preset travel path to the end point of the preset travel path within the first operation area; using the sensor module of the autonomous vehicle to perform point calibration on the preset travel path to obtain a calibrated position information set; performing position space judgment processing on the calibrated position information set to obtain the real-time operation position information of the autonomous vehicle within the first operation area; using a navigation error discrimination model to process the operation target path to generate the planned path information of the autonomous vehicle at the operation target point; driving the autonomous vehicle to travel according to the planned path information at the operation target point to reach the final target point. The present invention simultaneously uses three types of sensors, which can enable the autonomous vehicle to obtain accurate driving logic during operation, and has the characteristics of few interference factors, simple logic, and fast operation.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned vehicle navigation, and particularly to a fusion navigation method and device for an unmanned vehicle. Background Art

[0002] Currently, for the navigation of unmanned vehicles, it mainly includes means such as satellite navigation, laser navigation, and inertial navigation. For the laser navigation method, its cost is relatively high, and the requirements for the site are also relatively high. For the inertial navigation method, after long-term use, there will be serious cumulative errors. If not corrected in time, it will cause the unmanned vehicle to be unable to travel along the normal path. For the satellite navigation method, in signal shielding scenarios such as indoor warehouses and underground places, effective positioning cannot be achieved, which also affects the navigation and positioning effect.

[0003] On unmanned vehicles using multiple navigation methods, currently in the actual navigation process, only one navigation method is used at the same time, and the navigation information is not effectively fused. In a single environment scenario, a single navigation can obtain a relatively high navigation accuracy. However, in a scenario with large environmental differences, such as when the travel environment involves both indoor and outdoor at the same time, a single navigation method has problems of low navigation accuracy and poor applicability. Especially in a scenario with severe interference, it may even cause the single navigation method to fail and be unable to effectively navigate the unmanned vehicle. Summary of the Invention

[0004] To solve the problem that the existing unmanned vehicle navigation method only uses one navigation method at the same time and has a low navigation accuracy of the unmanned vehicle in a scenario with large environmental differences and severe interference, a first aspect of an embodiment of the present invention discloses a fusion navigation method for an unmanned vehicle, including:

[0005] S1, obtaining a preset running path and a first running area of the unmanned vehicle, and storing the preset running path and the first running area in the navigation module of the unmanned vehicle;

[0006] S2, driving the unmanned vehicle to travel along the preset running path to the end point of the preset running path in the first running area;

[0007] S3, during the process of the unmanned vehicle traveling along the preset running path, using the sensor module of the unmanned vehicle to calibrate the points on the preset running path to obtain a calibrated position information set; the calibrated position information set includes the position parameter information and the point arrival time information of several points collected by the sensor module; the sensor module of the unmanned vehicle includes a satellite positioning navigation sensor, a laser sensor, and an inertial navigation sensor;

[0008] S4, performing a position space judgment process on the calibrated position information set to obtain the real-time running position information of the unmanned vehicle in the first running area;

[0009] S5. Aggregate the real-time running positions to obtain the running target path of the driverless vehicle; the running target path includes several running target point position information.

[0010] S6. Process the running target path using the navigation error discrimination model to generate the planned path information of the driverless vehicle at the running target points; drive the driverless vehicle to travel according to the planned path information at the running target points and reach the final target point; the navigation error discrimination model includes a satellite navigation error discrimination model and a laser navigation error discrimination model.

[0011] As an optional implementation manner, in the first aspect of the embodiments of the present invention, during the process of the driverless vehicle traveling according to the preset running path, use the sensor module of the driverless vehicle to calibrate the points on the preset running path to obtain a set of calibration position information, including:

[0012] S31. During the process of the driverless vehicle traveling according to the preset running path, at preset time intervals, use the satellite positioning and navigation sensor, the laser sensor, and the inertial navigation sensor respectively to collect data on the position information of the point where the driverless vehicle is located, and obtain the position parameter information and the point arrival time information of the point where it is located; the position parameter information of the point where it is located includes satellite positioning parameters, laser sensor parameters, and inertial navigation parameters.

[0013] S32. Perform fusion processing on the position parameter information and the point arrival time information of the point where it is located to obtain a set of calibration position information.

[0014] As an optional implementation manner, in the first aspect of the embodiments of the present invention, the position space judgment processing of the set of calibration position information to obtain the real-time running position information of the driverless vehicle in the first running area includes:

[0015] S41. Judge the position parameter information of the point where it is located in the set of calibration position information. When the position parameter information is within the first running area, list the position parameter information as valid sensor position parameter information. When the position parameter information is outside the first running area, list the position parameter information as invalid sensor position parameter information.

[0016] S42. Perform weighted processing on the valid sensor position parameter information using the sensor confidence level; the sensor confidence level is determined according to the cumulative occurrence times of the invalid sensor position parameter information or the variance of the position parameter information.

[0017] S43. Use the point real-time position information and the point arrival time information to construct the real-time running position information of the driverless vehicle.

[0018] As an alternative implementation, in the first aspect of the embodiments of the present invention, the process of aggregating the real-time running positions to obtain the running target path of the driverless vehicle includes:

[0019] S51, sorting the real-time position information of the points according to the sequence of the arrival time information of the points to obtain the list information of the driverless vehicle points;

[0020] S52, using the list information of the driverless vehicle points to perform path planning on the driverless vehicle points in the list of the driverless vehicle points to obtain the running target path of the driverless vehicle.

[0021] As an alternative implementation, in the first aspect of the embodiments of the present invention, the process of using the navigation error discrimination model to process the running target path to generate the planned path information of the driverless vehicle at the running target point; driving the driverless vehicle to travel according to the planned path information at the running target point to reach the final target point includes:

[0022] S61, using the running target path to drive the driverless vehicle to the running target point;

[0023] S62, using the current position information of the driverless vehicle and the running target path to generate a set of driving information of the driverless vehicle at the target point;

[0024] S63, using the navigation error discrimination model to process the set of driving information of the target point to obtain the planned path information of the driverless vehicle at the running target point;

[0025] S64, driving the driverless vehicle to travel according to the planned path information at the running target point to reach the next running target point;

[0026] S65, judging the reached running target point. When the running target point is the final target point, the integrated navigation process of the driverless vehicle is completed; when the running target point is not the final target point, execute step S62.

[0027] As an alternative implementation, in the first aspect of the embodiments of the present invention, the process of using the current position information of the driverless vehicle and the running target path to generate a set of driving information of the driverless vehicle at the target point includes:

[0028] S621, using the sensor module to collect the current position information of the driverless vehicle;

[0029] S622. Establish a two-dimensional driving coordinate system with the center of the driverless vehicle as the origin, the direction of the vehicle's head as the X-axis, and the direction perpendicular to the vehicle's head as the Y-axis; the coordinates of the vehicle's head in the two-dimensional driving coordinate system are (X0, Y0).

[0030] S623. Determine that the coordinates of the next running target point of the driverless vehicle in the two-dimensional driving coordinate system are (X1, Y1) according to the current position information of the driverless vehicle and the running target path.

[0031] S624. Use the satellite positioning and navigation sensor to process the vehicle head coordinates (X0, Y0) and the running target point coordinates (X1, Y1) to generate the first driving vector coordinates (X2, Y2).

[0032] S625. Use the laser sensor to process the vehicle head coordinates (X0, Y0) and the running target point coordinates (X1, Y1) to generate the second driving vector coordinates (X3, Y3).

[0033] S626. Use the inertial navigation sensor to process the vehicle head coordinates (X0, Y0) and the running target point coordinates (X1, Y1) to generate the third driving vector coordinates (X4, Y4).

[0034] S627. Use the first driving vector, the second driving vector, and the third driving vector to construct the driving information set of the target point of the driverless vehicle.

[0035] As an optional implementation manner, in the first aspect of the embodiments of the present invention, the navigation error discrimination model includes a satellite navigation error discrimination model and a laser navigation error discrimination model, including:

[0036] The satellite navigation error discrimination model, whose expression is:

[0037] |X4 - X2| ≤ a,

[0038] |Y4 - Y2| ≤ a,

[0039] |(X2 - X1)(Y2 - Y1)| ≤ a2,

[0040] where the inertial navigation error threshold is a, and the satellite navigation error area constraint threshold is a2;

[0041] The laser navigation error discrimination model, whose expression is:

[0042] |X4 - X3| ≤ a,

[0043] |Y4 - Y3| ≤ a,

[0044] |(X3 - X1)(Y3 - Y1)| ≤ a4,

[0045] wherein, the laser navigation error area constraint threshold is a4.

[0046] As an alternative implementation, in the first aspect of the embodiments of the present invention, a navigation error discrimination model is used to process the target point driving information set to obtain the planned path information of the driverless vehicle at the operation target point, including:

[0047] S631, using the navigation error discrimination model to discriminate the target point driving information set; if the target point driving information set meets the satellite navigation error discrimination model, it is determined that the first driving vector is valid, if it does not meet the satellite navigation error discrimination model, it is determined that the first driving vector is invalid; if the target point driving information set meets the laser navigation error discrimination model, it is determined that the second driving vector is valid, if it does not meet the laser navigation error discrimination model, it is determined that the second driving vector is invalid;

[0048] S632, when only the third driving vector is valid, the third driving vector is used as the planned path information of the driverless vehicle at the operation target point; when only the first driving vector and the third driving vector are valid, the first driving vector is used as the planned path information of the driverless vehicle at the operation target point; when only the second driving vector and the third driving vector are valid, the second driving vector is used as the planned path information of the driverless vehicle at the operation target point; when the first driving vector, the second driving vector, and the third driving vector are all valid, the three driving vectors are processed using a weighting criterion to obtain the planned path information of the driverless vehicle at the operation target point.

[0049] The second aspect of the embodiments of the present invention discloses a fusion navigation device for a driverless vehicle, the device includes:

[0050] A memory storing executable program code;

[0051] A processor coupled to the memory;

[0052] The processor calls the executable program code stored in the memory to execute the driverless vehicle fusion navigation method described in the first aspect of the embodiments of the present invention.

[0053] The third aspect of the embodiments of the present invention discloses a computer storage medium storage, the computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the driverless vehicle fusion navigation method described in the first aspect of the embodiments of the present invention.

[0054] Beneficial effects

[0055] 1. The present invention provides a fusion navigation method for an unmanned vehicle. By simultaneously using three types of sensors during navigation, the unmanned vehicle for logistics handling can obtain precise driving logic during operation. It has the characteristics of fewer interference factors, simple logic, and fast operation, and has better applicability.

[0056] 2. During the execution of the steps of the method of the present invention, the running path of the unmanned vehicle is completed by selecting calibration points. And during the process of selecting calibration points, other unmanned vehicles can also be used as calibration points. Thus, during the operation of the unmanned vehicle, the handover of goods carried by the unmanned forklift or the handling of goods to multiple destinations in a single task can be completed, making the goods handling function of the unmanned vehicle more intelligent. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] 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.

[0058] Figure 1 is a schematic flow chart of the fusion navigation method for the unmanned vehicle of the present invention;

[0059] Figure 2 is a schematic diagram of the basic composition of the unmanned vehicle of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0061] Figure 1 is a schematic flow chart of the fusion navigation method for the unmanned vehicle of the present invention; Figure 2 is a schematic diagram of the basic composition of the unmanned vehicle of the present invention.

[0062] The following further describes the present invention with reference to the embodiments.

[0063] Embodiment 1

[0064] This embodiment discloses a fusion navigation method for an unmanned vehicle, including:

[0065] S1. Obtain the preset running path and the first running area of the driverless vehicle, and store the preset running path and the first running area in the navigation module of the driverless vehicle;

[0066] S2. Drive the driverless vehicle to travel along the preset running path to the end point of the preset running path within the first running area;

[0067] S3. During the process of the driverless vehicle traveling along the preset running path, use the sensor module of the driverless vehicle to calibrate the points on the preset running path to obtain a set of calibrated position information; the set of calibrated position information includes the position parameter information and the point arrival time information of several points collected by the sensor module; the sensor module of the driverless vehicle includes a satellite positioning and navigation sensor, a laser sensor, and an inertial navigation sensor;

[0068] S4. Perform a position space judgment process on the set of calibrated position information to obtain the real-time running position information of the driverless vehicle within the first running area;

[0069] S5. Perform an aggregation process on the real-time running position to obtain the running target path of the driverless vehicle; the running target path includes several pieces of running target point information;

[0070] S6. Use the navigation error discrimination model to process the running target path to generate the planned path information of the driverless vehicle at the running target point; drive the driverless vehicle to travel according to the planned path information at the running target point to reach the final target point; the navigation error discrimination model includes a satellite navigation error discrimination model and a laser navigation error discrimination model.

[0071] Set a certain number of laser-irradiated reflectors on the preset running path of the driverless vehicle. The laser sensor can obtain the angle and distance information between the driverless vehicle and the laser-irradiated reflectors by irradiating the laser-irradiated reflectors.

[0072] During the process of the driverless vehicle traveling along the preset running path, using the sensor module of the driverless vehicle to calibrate the points on the preset running path to obtain a set of calibrated position information includes:

[0073] S31. During the process of the driverless vehicle traveling along the preset running path, at preset time intervals, use the satellite positioning and navigation sensor, the laser sensor, and the inertial navigation sensor respectively to collect the position information of the point where the driverless vehicle is located, and obtain the position parameter information and the point arrival time information of the point where it is located; the position parameter information of the point where it is located includes satellite positioning parameters, laser sensor parameters, and inertial navigation parameters;

[0074] S32. Perform fusion processing on the position parameter information and the point arrival time information of the current position to obtain a calibrated position information set.

[0075] The step S32 includes:

[0076] Use the position parameter information of a current position and the corresponding point arrival time information to fuse and form a set of position, and all the sets of position constitute the calibrated position information set.

[0077] The performing position space judgment processing on the calibrated position information set to obtain the real-time running position information of the driverless vehicle in the first running area includes:

[0078] S41. Judge the position parameter information of the current position in the calibrated position information set. When the position parameter information is within the first running area, list the position parameter information as valid sensor position parameter information; when the position parameter information is outside the first running area, list the position parameter information as invalid sensor position parameter information;

[0079] S42. Perform weighted processing on the valid sensor position parameter information by using the sensor confidence level to obtain the real-time position information of the point; the sensor confidence level is determined according to the cumulative occurrence times of the invalid sensor position parameter information or the variance of the position parameter information;

[0080] The sensor confidence level is determined according to the cumulative occurrence times of the invalid sensor position parameter information, including:

[0081] After judging the position parameter information of the current position in the calibrated position information set, for each type of sensor, update the cumulative occurrence times of its corresponding invalid sensor position parameter information to obtain the cumulative occurrence times of the invalid sensor position parameter information of each type of sensor, and perform normalization processing on the cumulative occurrence times. The obtained result is used as the sensor confidence level of this type of sensor.

[0082] The sensor confidence level is determined according to the variance of the position parameter information, including:

[0083] After judging the position parameter information of the current position in the calibrated position information set, for each type of sensor, calculate the variance value of all its valid sensor position parameter information, and perform normalization processing on the variance value. The obtained result is used as the sensor confidence level of this type of sensor.

[0084] S43. Use the real-time position information of the point and the point arrival time information to construct the real-time running position information of the driverless vehicle.

[0085] The step S43 includes:

[0086] Using the real-time position information of the points and the corresponding arrival time information of the points, a real-time position group is formed; a set is constructed using all the real-time position groups, and the set is the real-time running position information of the driverless vehicle.

[0087] Performing a pooling process on the real-time running position to obtain the running target path of the driverless vehicle, including:

[0088] S51, sorting the real-time position information of the points according to the sequence of the arrival time information of the points to obtain the driverless vehicle point list information;

[0089] S52, using the driverless vehicle point list information to perform path planning on the driverless vehicle points in the driverless vehicle point list to obtain the running target path of the driverless vehicle.

[0090] The step S52 includes:

[0091] Using the driverless vehicle point list information to connect the driverless vehicle points in the order of their appearance to obtain a driverless vehicle running path, and taking the driverless vehicle running path as the running target path of the driverless vehicle.

[0092] Processing the running target path using the navigation error discrimination model to generate the planned path information of the driverless vehicle at the running target point; driving the driverless vehicle to travel according to the planned path information at the running target point to reach the final target point, including:

[0093] S61, driving the driverless vehicle to the running target point using the running target path;

[0094] S62, generating a set of driving information of the driverless vehicle at the target point using the current position information of the driverless vehicle and the running target path;

[0095] S63, processing the set of driving information of the driverless vehicle at the target point using the navigation error discrimination model to obtain the planned path information of the driverless vehicle at the running target point;

[0096] S64, driving the driverless vehicle to travel according to the planned path information at the running target point to reach the next running target point;

[0097] S65, judging the reached running target point. When the running target point is the final target point, the fusion navigation process of the driverless vehicle is completed; when the running target point is not the final target point, execute step S62.

[0098] In step S64, when driving the driverless vehicle to travel according to the planned path information of the operating target point, the position information of the vehicle is collected in real time by using sensors, and the position information is compared with the driverless vehicle points in the operating target path. When the distance between the position information and the driverless vehicle points is less than a preset threshold, it is determined that the driverless vehicle has reached the next operating target point;

[0099] Generating a set of target point driving information for the driverless vehicle by using the current position information of the driverless vehicle and the operating target path, including:

[0100] S621, collecting the current position information of the driverless vehicle by using the sensor module;

[0101] S622, taking the center of the driverless vehicle as the origin, taking the head direction of the driverless vehicle as the X-axis, and taking the direction perpendicular to the head as the Y-axis to establish a two-dimensional driving coordinate system; the coordinates of the head of the driverless vehicle in the two-dimensional driving coordinate system are (X0, Y0);

[0102] S623, determining that the coordinates of the next operating target point of the driverless vehicle in the two-dimensional driving coordinate system are (X1, Y1) according to the current position information of the driverless vehicle and the operating target path;

[0103] S624, processing the head coordinates (X0, Y0) of the driverless vehicle and the coordinates (X1, Y1) of the operating target point by using the satellite positioning and navigation sensor to generate a first driving vector coordinate (X2, Y2); the driving vector refers to the amount of movement required for the driverless vehicle to reach the operating target point when navigating by using the corresponding sensor module. For example, the driving vector coordinate of the satellite positioning and navigation sensor is (X2, Y2), which means that when the driverless vehicle navigates by using the satellite positioning and navigation sensor, it travels X2 along the X-axis and Y2 along the Y-axis to reach the operating target point.

[0104] S625, processing the head coordinates (X0, Y0) of the driverless vehicle and the coordinates (X1, Y1) of the operating target point by using the laser sensor to generate a second driving vector coordinate (X3, Y3);

[0105] S626, processing the head coordinates (X0, Y0) of the driverless vehicle and the coordinates (X1, Y1) of the operating target point by using the inertial navigation sensor to generate a third driving vector coordinate (X4, Y4);

[0106] S627, constructing a set of target point driving information for the driverless vehicle by using the first driving vector, the second driving vector, and the third driving vector.

[0107] Construct a set of driving vectors using the first driving vector, the second driving vector, and the third driving vector, and use the set of driving vectors as the target point driving information set of the driverless vehicle.

[0108] The navigation error discrimination model includes a satellite navigation error discrimination model and a laser navigation error discrimination model, and includes:

[0109] The satellite navigation error discrimination model has the following expression:

[0110] |X4 - X2| ≤ a,

[0111] |Y4 - Y2| ≤ a,

[0112] |(X2 - X1)(Y2 - Y1)| ≤ a2,

[0113] where the inertial navigation error threshold is a, and the satellite navigation error area constraint threshold is a2;

[0114] The laser navigation error discrimination model has the following expression:

[0115] |X4 - X3| ≤ a,

[0116] |Y4 - Y3| ≤ a,

[0117] |(X3 - X1)(Y3 - Y1)| ≤ a4,

[0118] where the laser navigation error area constraint threshold is a4.

[0119] Using the navigation error discrimination model to process the target point driving information set to obtain the planned path information of the driverless vehicle at the operation target point, including:

[0120] S631. Use the navigation error discrimination model to discriminate the target point driving information set; if the target point driving information set meets the satellite navigation error discrimination model, determine that the first driving vector is valid, if it does not meet the satellite navigation error discrimination model, determine that the first driving vector is invalid; if the target point driving information set meets the laser navigation error discrimination model, determine that the second driving vector is valid, if it does not meet the laser navigation error discrimination model, determine that the second driving vector is invalid;

[0121] S632, when only the third driving vector is valid, use the third driving vector as the planned path information of the driverless vehicle at the operation target point; when only the first driving vector and the third driving vector are valid, use the first driving vector as the planned path information of the driverless vehicle at the operation target point; when only the second driving vector and the third driving vector are valid, use the second driving vector as the planned path information of the driverless vehicle at the operation target point; when the first driving vector, the second driving vector, and the third driving vector are all valid, process the three driving vectors using a weighting criterion to obtain the planned path information of the driverless vehicle at the operation target point.

[0122] The process of using the weighting criterion to process the three driving vectors to obtain the planned path information of the driverless vehicle at the operation target point includes:

[0123] Determine the proportional value of the weight of the first driving vector and the weight of the second driving vector using the current number of service satellites of the satellite positioning and navigation sensor and the number of laser-irradiated reflectors set at the operation target point;

[0124] Determine the weight of the third driving vector using the value range of the sum of the current number of service satellites of the satellite positioning and navigation sensor and the number of laser-irradiated reflectors set at the operation target point;

[0125] Solve for the weight of the first driving vector and the weight of the second driving vector using the constraint condition that the sum of the weights of the three driving vectors is 1;

[0126] Perform a weighted summation operation on the three driving vectors using the weights of the three driving vectors to obtain the planned path information of the driverless vehicle at the operation target point.

[0127] The process of using the weighting criterion to process the three driving vectors to obtain the planned path information of the driverless vehicle at the operation target point includes:

[0128] The calculation formula of the weighting criterion is:

[0129] X1 = X2×ik + X3×k + X4×p,

[0130] Y1 = Y2×ik + Y3×k + Y4×p,

[0131] where k is the weighting coefficient of the second driving vector, i is the ratio of the weighting coefficients of the first driving vector and the second driving vector, and p is the weighting coefficient of the third driving vector, which is determined by the current number of service satellites of the satellite positioning and navigation sensor and the number of laser-irradiated reflectors set at the operation target point. The determination process is shown in Table 1.

[0132] Table 1 Determination Table of Weighting Coefficient Ratio

[0133]

[0134] When the sum of the current number of service satellites of the satellite positioning and navigation sensor and the number of laser-irradiated reflectors set at the operating target point is greater than or equal to 6, p = 0.1; when the sum of the current number of service satellites of the satellite positioning and navigation sensor and the number of laser-irradiated reflectors set at the operating target point is less than 6, and the sum is represented as y, then the value of p is 1 - 0.15p.

[0135] One of the main application scenarios of the driverless vehicle is automatic logistics handling and transportation. AGV is the abbreviation of Automated Guided Vehicle, which means "automated guided transport vehicle". An AGV driverless vehicle is a transport vehicle equipped with automatic guiding devices such as electromagnetic or optical ones, capable of traveling along a specified guiding path, and having safety protection and various loading and unloading functions. The AGV driverless vehicle automatically transports items to a designated location through navigation. The most common navigation methods include magnetic stripe guidance, laser guidance, ultra-high frequency RFID guidance, etc. For the magnetic stripe navigation method, its site setting has certain limitations and has a certain impact on the site decoration style. For the ultra-high frequency RFID navigation method, its cost is moderate. Its advantages are high guiding accuracy, more convenient site setting, which can meet complex site layout scenarios, and there is no need to change the overall decoration environment of the place. At the same time, the high safety and stability of the ultra-high frequency RFID navigation method are also not possessed by the magnetic stripe guidance and laser navigation methods.

[0136] This embodiment provides a fusion navigation method for a driverless vehicle. By simultaneously using three types of sensors during the navigation process, the driverless vehicle for logistics handling can obtain accurate driving logic during operation, which has the characteristics of fewer interference factors, simple logic, and fast operation, and has better applicability.

[0137] In the process of executing the steps of the method disclosed in this embodiment, the running path of the driverless vehicle is completed through the selection of calibration points. And during the process of selecting calibration points, other driverless vehicles can also be used as calibration points. Thus, during the running process of the driverless vehicle, the goods transfer of the unmanned forklift or the goods handling work with multiple destinations for a single task can be completed, making the goods handling function of the driverless vehicle more intelligent.

[0138] Embodiment 2

[0139] This embodiment discloses a fusion navigation device for a driverless vehicle. The device includes:

[0140] A memory storing executable program code;

[0141] A processor coupled to the memory;

[0142] The processor calls the executable program code stored in the memory and executes the unmanned vehicle fusion navigation method described in Embodiment 1.

[0143] Embodiment 3

[0144] This embodiment discloses a computer storage medium storing computer instructions that, when called, are used to execute the unmanned vehicle fusion navigation method described in Embodiment 1.

[0145] Embodiment 4

[0146] This embodiment discloses a fusion navigation method for an unmanned vehicle, which includes sensors, a navigation module, a travel controller, a motor, etc.; the sensors, navigation module, travel controller, and motor are connected in sequence; the sensors include a satellite positioning and navigation sensor, a laser sensor, and an inertial navigation sensor. The navigation module constitutes the control system of the unmanned vehicle; the travel controller and the motor constitute the travel system of the unmanned vehicle, and the composition of the unmanned vehicle is as Figure 2 shown; the fusion navigation method of the unmanned vehicle includes:

[0147] First, perform path calibration. The specific implementation is to manually drive the unmanned vehicle to calibrate the target path, mark each point of the calibrated path, and record the parameters of each sensor at each point as follows:

[0148] Point 1 (laser sensor point parameters (including at least three reflector measurement data, angle and distance differences), differential satellite sensor point parameters, inertial navigation point parameters);

[0149] Point 2 (laser sensor point parameters, differential satellite sensor point parameters, inertial navigation point parameters);

[0150] ······

[0151] Point x (laser sensor point parameters, differential satellite sensor point parameters, inertial navigation point parameters);

[0152] The calibration generates data for each point, stores it in the navigation module, and simultaneously transmits it to the scheduling system;

[0153] Through the path position calibration of the unmanned vehicle's operation and the electronic map stored in the navigation hardware, confirm the real-time positions of each unmanned vehicle in the unmanned vehicle's operation area;

[0154] Perform aggregation processing on the real-time positions to obtain the operation target path of the unmanned vehicle; the operation target path includes several operation target point information;

[0155] Process the running target path using a navigation error discrimination model to generate the planned path information of the driverless vehicle at the running target point; drive the driverless vehicle to travel according to the planned path information at the running target point to reach the final target point; the navigation error discrimination model includes a satellite navigation error discrimination model and a laser navigation error discrimination model.

[0156] Construct a database, name each driverless vehicle in the running area of the driverless vehicle, establish a data storage subfolder in the constructed database according to the name of the driverless vehicle, and store the running data of each driverless vehicle in real-time and distinguish them.

[0157] Before being deployed on the driverless vehicle, the navigation hardware is used to analyze the path image of the running area of the driverless vehicle, analyze the feasible path of the driverless vehicle in the running area of the driverless vehicle in the path image, and design an electronic map according to the feasible path of the driverless vehicle and upload it to the navigation hardware.

[0158] When driving the driverless vehicle to run according to the data stored in the navigation hardware, a calibration period and a calibration distance are synchronously set, and the running path of the driverless vehicle is calibrated during the running process of the driverless vehicle according to the calibration period; wherein, the calibration distance is set to 1.5 - 2m.

[0159] When calibrating the running path of the driverless vehicle, the satellite navigation positioning sensor, laser sensor and inertial navigation sensor deployed on the driverless vehicle are applied. According to the calibration period, the parameters captured by each sensor when the calibration period changes are used as calibration data and stored in the driverless vehicle.

[0160] The satellite navigation positioning sensor has a differential function.

[0161] When the data storage subfolder receives new running data of the driverless vehicle each time, it compares the newly received running data of the driverless vehicle with the existing data stored in the data storage subfolder to identify whether there is data with an opposite running logic to the newly received running data of the driverless vehicle in the existing data stored in the data storage subfolder; when the identification result is yes, delete the newly received running data of the driverless vehicle and the existing data stored in the data storage subfolder and having an opposite running logic to the newly received running data of the driverless vehicle, and when the identification result is no, send the newly received running data of the driverless vehicle to the data storage subfolder and store it.

[0162] In summary, through the execution of the steps in the method, the driverless vehicle for logistics handling can obtain accurate driving logic during operation. It has the characteristics of few interference factors, simple logic, and fast operation, with good applicability. Moreover, during the execution of the steps of the method, the running path of the driverless vehicle is completed by selecting calibration points. And during the process of selecting calibration points, other driverless vehicles can also be used as calibration points. Thus, during the operation of the driverless vehicle, it can complete the handover of goods carried by the driverless vehicle or the handling of goods to multiple destinations in a single task, making the goods handling function of the driverless vehicle more intelligent.

[0163] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements will 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 invention.

Claims

1. A fusion navigation method for an autonomous vehicle, characterized in that, Including: S1. Obtain the preset running path and the first running area of the driverless vehicle, and store the preset running path and the first running area in the navigation module of the driverless vehicle; S2. Drive the driverless vehicle to travel along the preset running path to the end of the preset running path within the first running area; S3. During the process of the driverless vehicle traveling along the preset running path, use the sensor module of the driverless vehicle to calibrate the points on the preset running path to obtain a set of calibrated position information; The set of calibrated position information includes the position parameter information and the point arrival time information of several points collected by the sensor module; the sensor module of the driverless vehicle includes a satellite positioning and navigation sensor, a laser sensor, and an inertial navigation sensor; S4. Perform position space judgment processing on the set of calibrated position information to obtain the real-time running position information of the driverless vehicle within the first running area; S5. Perform aggregation processing on the real-time running position to obtain the running target path of the driverless vehicle; The running target path includes several running target point information; S6. Use the navigation error discrimination model to process the running target path to generate the planned path information of the driverless vehicle at the running target point; drive the driverless vehicle to travel according to the planned path information at the running target point to reach the final target point; the navigation error discrimination model includes a satellite navigation error discrimination model and a laser navigation error discrimination model; Using the navigation error discrimination model to process the running target path to generate the planned path information of the driverless vehicle at the running target point; Driving the driverless vehicle to travel according to the planned path information at the running target point to reach the final target point, including: S61. Use the running target path to drive the driverless vehicle to the running target point; S62. Use the current position information of the driverless vehicle and the running target path to generate a set of target point travel information of the driverless vehicle; S63. Use the navigation error discrimination model to process the set of target point travel information to obtain the planned path information of the driverless vehicle at the running target point; S64. Drive the driverless vehicle to travel according to the planned path information at the running target point to reach the next running target point; S65. Judge the reached running target point. When the running target point is the final target point, complete the integrated navigation process of the driverless vehicle; when the running target point is not the final target point, execute step S62.

2. The fusion navigation method for an autonomous vehicle according to claim 1, characterized in that, During the process of the driverless vehicle traveling along the preset running path, using the sensor module of the driverless vehicle to calibrate the points on the preset running path to obtain a set of calibrated position information, including: S31. During the process of the driverless vehicle traveling along the preset running path, at preset time intervals, respectively use the satellite positioning and navigation sensor, the laser sensor, and the inertial navigation sensor to collect data on the position information of the point where the driverless vehicle is located to obtain the position parameter information and the point arrival time information of the point where it is located; the position parameter information of the point where it is located includes satellite positioning parameters, laser sensor parameters, and inertial navigation parameters; S32. Fuse the position parameter information and the point arrival time information of the current position to obtain a calibrated position information set.

3. The fusion navigation method for an autonomous vehicle according to claim 1, characterized in that, The position space judgment process for the calibrated position information set to obtain the real-time running position information of the driverless vehicle in the first operation area includes: S41. Judge the position parameter information of the current position in the calibrated position information set. When the position parameter information is within the first operation area, list the position parameter information as valid sensor position parameter information; when the position parameter information is outside the first operation area, list the position parameter information as invalid sensor position parameter information. S42. For the valid sensor position parameter information, perform weighted processing using the sensor confidence level to obtain the real-time position information of the point. The sensor confidence level is determined according to the cumulative occurrence times of the invalid sensor position parameter information or the variance of the position parameter information. S43. Use the real-time position information of the point and the point arrival time information to construct the real-time running position information of the driverless vehicle.

4. The fusion navigation method for an autonomous vehicle according to claim 3, characterized in that, The aggregation process for the real-time running position to obtain the running target path of the driverless vehicle includes: S51. Sort the real-time position information of the points in the order of the point arrival time information to obtain the driverless vehicle point list information. S52. Use the driverless vehicle point list information to perform path planning for the driverless vehicle points in the driverless vehicle point list to obtain the running target path of the driverless vehicle.

5. The fusion navigation method for an autonomous vehicle according to claim 1, characterized in that, The generation of the target point driving information set of the driverless vehicle using the current position information of the driverless vehicle and the running target path includes: S621. Use the sensor module to collect the current position information of the driverless vehicle. S622. Establish a two-dimensional driving coordinate system with the center of the driverless vehicle as the origin, the direction of the driverless vehicle's head as the X-axis, and the direction perpendicular to the head as the Y-axis. The coordinates of the driverless vehicle's head in the two-dimensional driving coordinate system are (X0, Y0). S623. According to the current position information of the driverless vehicle and the running target path, determine that the coordinates of the next running target point of the driverless vehicle in the two-dimensional driving coordinate system are (X1, Y1). S624. Use the satellite positioning and navigation sensor to process the driverless vehicle head coordinates (X0, Y0) and the running target point coordinates (X1, Y1) to generate the first driving vector coordinates (X2, Y2). S625. Use the laser sensor to process the driverless vehicle head coordinates (X0, Y0) and the running target point coordinates (X1, Y1) to generate the second driving vector coordinates (X3, Y3). S626. Use the inertial navigation sensor to process the driverless vehicle head coordinates (X0, Y0) and the running target point coordinates (X1, Y1) to generate the third driving vector coordinates (X4, Y4). S627. Use the first driving vector, the second driving vector, and the third driving vector to construct the target point driving information set of the driverless vehicle.

6. The integrated navigation method for an autonomous vehicle according to claim 5, wherein the navigation error discrimination model includes a satellite navigation error discrimination model and a laser navigation error discrimination model, and comprises: The satellite navigation error discrimination model has the following expression: |X4 - X2| ≤ a, |Y4 - Y2| ≤ a, |(X2 - X1)(Y2 - Y1)| ≤ a2, where the inertial navigation error threshold is a, and the satellite navigation error area constraint threshold is a2; The laser navigation error discrimination model has the following expression: |X4 - X3| ≤ a, |Y4 - Y3| ≤ a, |(X3 - X1)(Y3 - Y1)| ≤ a4, where the laser navigation error area constraint threshold is a4.

7. The integrated navigation method for an autonomous vehicle according to claim 5, characterized in that Using the navigation error discrimination model to process the target point driving information set, the planned path information of the driverless vehicle at the operating target point is obtained, including: S631. Using the navigation error discrimination model to discriminate the target point driving information set; if the target point driving information set satisfies the satellite navigation error discrimination model, it is determined that the first driving vector is valid, and if it does not satisfy the satellite navigation error discrimination model, it is determined that the first driving vector is invalid; if the target point driving information set satisfies the laser navigation error discrimination model, it is determined that the second driving vector is valid, and if it does not satisfy the laser navigation error discrimination model, it is determined that the second driving vector is invalid; S632. When only the third driving vector is valid, the third driving vector is used as the planned path information of the driverless vehicle at the operating target point; when only the first driving vector and the third driving vector are valid, the first driving vector is used as the planned path information of the driverless vehicle at the operating target point; when only the second driving vector and the third driving vector are valid, the second driving vector is used as the planned path information of the driverless vehicle at the operating target point; when the first driving vector, the second driving vector, and the third driving vector are all valid, the three driving vectors are processed using a weighting criterion to obtain the planned path information of the driverless vehicle at the operating target point.

8. An integrated navigation device for an autonomous vehicle, characterized in that The device includes: A memory storing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory and executes the fusion navigation method of the driverless vehicle according to any one of claims 1 - 7.

9. A computer storage medium, characterized in that The computer storage medium stores computer instructions, which are used to execute the fusion navigation method of the driverless vehicle according to any one of claims 1 - 7 when called.

Images