An automatic driving vehicle multi-sensor optimization configuration method and system

Abstract

The application provides a kind of automatic driving car multi-sensor optimization configuration method and system, belongs to automatic driving car technical field, obtains the structural and size parameter information of the automatic driving car to be configured;According to the model of automatic driving car, the perception area of the automatic driving car to be configured is determined, including core perception area and extended perception area;According to the parameter information of the automatic driving car to be configured obtained, in combination with the perception area of automatic driving car, the function and performance parameters of different types of sensors are considered, and the feasible arrangement area of each type of sensor is determined according to the perception area discretization mechanism;With the minimum sensor arrangement cost, the maximum coverage and redundancy range as the target, a mixed integer programming model is constructed;Solve the mixed integer programming model to obtain the optimal sensor quantity and installation position.The application reduces the cost while ensuring the coverage and redundancy, improves the safety and reliability of the whole intelligent driving system.

Description

Technical Field

[0001] This invention relates to the field of autonomous vehicle technology, and specifically to a method and system for optimizing the configuration of multiple sensors in an autonomous vehicle. Background Technology

[0002] Environmental perception, a key technology for autonomous vehicles, utilizes sensors such as LiDAR, cameras, millimeter-wave radar, and ultrasonic radar to perceive the surrounding environment. It dynamically monitors changes in the environment in real time and makes decisions based on the acquired information, enabling boundary monitoring, vehicle detection, pedestrian detection, and supporting path planning and navigation decisions for autonomous vehicles. With the development of machine learning, especially the rise of deep learning technology, environmental perception has received widespread attention from industry and academia. Currently, there are two main approaches to environmental perception technology: one is a multi-sensor fusion scheme primarily based on cameras; the other is a scheme primarily based on LiDAR, supplemented by other sensors. Different sensors have their own advantages and limitations in terms of perception range. With the maturity of LiDAR technology and its significant cost reduction, LiDAR-led perception schemes have gained increasing attention due to their advantages such as higher resolution, stronger resistance to active interference, and good low-altitude detection performance. How to rationally deploy multiple sensors and use sensor information fusion technology to compensate for the shortcomings of single-sensor systems, thereby improving the safety and reliability of the entire intelligent driving system, has significant theoretical and practical implications. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for optimizing the configuration of multiple sensors in autonomous vehicles while reducing costs, ensuring coverage and redundancy, and improving the safety and reliability of the entire intelligent driving system, so as to solve at least one of the technical problems existing in the background art.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] On one hand, the present invention provides a method for optimizing the configuration of multiple sensors in an autonomous vehicle, comprising:

[0006] Obtain structural and dimensional parameters of the autonomous vehicle to be configured;

[0007] Based on the model of the autonomous vehicle, determine the perception area (RoIs) of the autonomous vehicle to be configured, including the core perception area and the extended perception area;

[0008] Based on the parameter information of the autonomous vehicle to be configured, combined with the perception area of ​​the autonomous vehicle, and considering the functions and performance parameters of different types of sensors, the feasible placement areas of various sensors are determined according to the perception area discretization mechanism.

[0009] A mixed-integer programming model is constructed with the goal of minimizing sensor deployment cost, maximizing coverage and redundancy area.

[0010] Solve the mixed integer programming model to obtain the optimal number of sensors and their installation locations.

[0011] Preferably, the parameter information of the autonomous vehicle to be configured includes: the size parameters of the autonomous vehicle to be configured and the parameters of the sensors to be deployed; the parameters of the sensors to be deployed include: the horizontal field of view (FoV) and range of the solid-state lidar, the horizontal field of view and range of the camera, the horizontal field of view and range of the millimeter-wave radar, and the horizontal field of view and range of the ultrasonic radar.

[0012] Preferably, a mixed-integer programming model is constructed, including:

[0013] Four binary decision variables were defined to determine the placement of the four types of sensors:

[0014]

[0015] If a solid-state LiDAR is deployed at location s, then α s =1, otherwise α s =0;

[0016]

[0017] If a camera is placed at position c, then β c =1, otherwise β c =0;

[0018]

[0019] If a millimeter-wave radar is deployed at location m, then χ m =1, otherwise χ m =0;

[0020]

[0021] If an ultrasonic radar is deployed at position u, then δ u =1, otherwise δu =0;

[0022] Construct a multi-objective optimization function:

[0023] O=min(ψ1O1+ψ2O2+ψ3O3) (5)

[0024] Where O1, O2, and O3 represent cost, coverage, and redundancy, respectively, and ψ1, ψ2, and ψ3 represent the weights of the three sub-objectives;

[0025] The cost is determined by the price of the sensor, as defined below:

[0026]

[0027] in and ω1, ω2, ω3, and ω4 represent the total number of solid-state LiDAR, cameras, millimeter-wave radar, and ultrasonic radar, respectively, while ω4 represent the prices of the four types of sensors: solid-state LiDAR, cameras, millimeter-wave radar, and ultrasonic radar.

[0028] Preferably, the perception area of ​​the autonomous vehicle to be configured is divided into a core perception area and an extended perception area. The coverage rate is defined as the weighted sum of the coverage rates of the core perception area and the extended perception area, as follows:

[0029]

[0030] Where S v1 and S v2 σ1 and σ2 respectively represent the coverage area of ​​the effective core sensing area and the area of ​​the extended sensing area, and represent the importance of the two types of coverage rates;

[0031] N sum The effective coverage area of ​​each sensor is defined as follows:

[0032]

[0033] Redundancy is the ratio of the total sensing area to the redundant coverage area of ​​multiple source sensors, defined as follows:

[0034]

[0035] Where S core ∪S extend S represents the area of ​​the perception zone for an autonomous vehicle. g and S h S represents the coverage area of ​​the sensor g and h. g ∩S h Corresponding to the area of ​​the redundant region covered by the two sensors, This represents the sum of the effective redundant areas covered by all sensors.

[0036] Preferably, the mixed-integer programming model is constructed to satisfy the following constraints:

[0037] S core ×γ1≤S v1 (9)

[0038] Where γ1∈[0,1], the effective coverage core area of ​​the multi-source sensor cannot be less than γ1 times the core sensing area;

[0039] S extend ×γ2≤S v2 (10)

[0040] Where γ2∈[0,1], the effective coverage extension area of ​​the multi-source sensor cannot be less than γ2 times the extended sensing area;

[0041]

[0042] Where γ∈[0,1], the redundancy of the sensing area cannot be less than γ times the total sensing area;

[0043]

[0044] Where N S The integer is N. Considering the high cost of solid-state lidar, its quantity will not exceed N. S indivual;

[0045]

[0046] in and The integer is positive, and considering the possible placement locations of millimeter-wave radars, its number is in Within the range;

[0047]

[0048] in and The number of cameras is a positive integer, taking into account the possible placement locations of the cameras. Within the range;

[0049]

[0050] in and The number of ultrasonic radars is a positive integer, taking into account the possible placement locations. Within the range;

[0051] α s ∈{0,1},βc ∈{0,1},χ m ∈{0,1},δ u ∈{0,1} (16)

[0052] Ensure that the above variables are binary 0-1 variables.

[0053] Preferably, solving the mixed integer programming model includes: setting the parameters in the objective function and constraints, solving the model based on the genetic simulated annealing algorithm and CPLEX, and outputting the optimal configuration scheme for multiple sensors.

[0054] Secondly, the present invention provides a multi-sensor optimized configuration system for autonomous vehicles, comprising:

[0055] The acquisition module is used to acquire parameter information such as the structure and size of the autonomous vehicle to be configured;

[0056] The perception area segmentation module is used to determine the perception area of ​​the autonomous vehicle to be configured, including the core perception area and the extended perception area, based on the vehicle model.

[0057] The feasible deployment area construction module is used to determine the feasible deployment areas of various sensors based on the parameter information of the autonomous vehicle to be configured, combined with the perception area of ​​the autonomous vehicle, considering the functional and performance parameters of different types of sensors, and according to the perception area discretization mechanism.

[0058] The planning model building module is used to construct a mixed-integer programming model with the goal of minimizing sensor deployment cost, maximizing coverage and redundancy area.

[0059] The solution module is used to solve the mixed integer programming model to obtain the optimal number of sensors and their installation locations.

[0060] The beneficial effects of this invention are as follows: Based on the given vehicle structure and dimensions, the types of solid-state LiDAR, cameras, millimeter-wave radar, and ultrasonic radar are selected, and parameters such as field of view and distance are determined. The perception area of ​​the autonomous vehicle is divided into a core perception area and an extended perception area, which are then discretized. Based on the functions and perception ranges of various sensors, feasible sensor placement areas are set. Considering cost, coverage, and redundancy, a quantitative optimization configuration scheme for multi-source sensors is determined. A mixed-integer programming model is constructed and solved using the genetic simulated annealing algorithm and CPLEX. The sensor placement positions and quantities are optimized, improving coverage and redundancy while reducing costs, thus obtaining the optimal configuration scheme.

[0061] The advantages of additional aspects of the invention will be set forth more clearly in the following description or will be learned by practice of the invention. Attached Figure Description

[0062] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0063] Figure 1 This is a flowchart of the multi-sensor optimization configuration method for autonomous vehicles according to an embodiment of the present invention.

[0064] Figure 2 This is a schematic diagram of the sensing area of ​​various sensors described in the embodiments of the present invention.

[0065] Figure 3 This is a schematic diagram illustrating the division of the perception area of ​​an autonomous vehicle according to an embodiment of the present invention.

[0066] Figure 4 This is a schematic diagram illustrating the calculation of the effective perception area of ​​an autonomous vehicle according to an embodiment of the present invention.

[0067] Figure 5 This is a schematic diagram illustrating the feasible region division of the sensor according to an embodiment of the present invention.

[0068] Figure 6 This is a schematic diagram of the multi-source sensor optimization configuration scheme described in an embodiment of the present invention. Detailed Implementation

[0069] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0070] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0071] It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless defined as here.

[0072] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, and / or groups thereof.

[0073] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0074] To facilitate understanding of the present invention, the present invention will be further explained and described below with reference to the accompanying drawings and specific embodiments, and the specific embodiments do not constitute a limitation on the embodiments of the present invention.

[0075] Those skilled in the art should understand that the accompanying drawings are merely schematic diagrams of embodiments, and the components in the drawings are not necessarily essential for implementing the present invention.

[0076] Example 1

[0077] This embodiment 1 provides a multi-sensor optimized configuration system for autonomous vehicles, including:

[0078] The acquisition module is used to acquire parameter information such as the structure and size of the autonomous vehicle to be configured;

[0079] The perception area segmentation module is used to determine the perception area of ​​the autonomous vehicle to be configured, including the core perception area and the extended perception area, based on the vehicle model.

[0080] The feasible deployment area construction module is used to determine the feasible deployment areas of various sensors based on the parameter information of the autonomous vehicle to be configured, combined with the perception area of ​​the autonomous vehicle, considering the functional and performance parameters of different types of sensors, and according to the perception area discretization mechanism.

[0081] The planning model building module is used to construct a mixed-integer programming model with the goal of minimizing sensor deployment cost, maximizing coverage and redundancy area.

[0082] The solution module is used to solve the mixed integer programming model to obtain the optimal number of sensors and their installation locations.

[0083] In this embodiment 1, the above-described system is used to implement a multi-sensor optimization configuration method for autonomous vehicles, including:

[0084] The acquisition module is used to obtain parameter information such as the structure and size of the autonomous vehicle to be configured;

[0085] The perception area segmentation module is used to determine the perception area of ​​the autonomous vehicle to be configured, including the core perception area and the extended perception area, based on the vehicle model.

[0086] The feasible deployment area construction module is used to determine the feasible deployment areas of various sensors based on the acquired parameter information of the autonomous vehicle to be configured, combined with the perception area of ​​the autonomous vehicle, considering the functional and performance parameters of different types of sensors, and according to the perception area discretization mechanism.

[0087] Using a planning model building module, a mixed-integer programming model is constructed with the goal of minimizing sensor deployment cost, maximizing coverage and redundancy area.

[0088] The mixed integer programming model is solved using the solver module to obtain the optimal number of sensors and their installation locations.

[0089] The parameter information of the autonomous vehicle to be configured includes: the size parameters of the autonomous vehicle to be configured and the parameters of the sensors to be deployed; the parameters of the sensors to be deployed include: the horizontal field of view and distance of the solid-state lidar, the horizontal field of view and distance of the camera, the horizontal field of view and distance of the millimeter-wave radar, and the horizontal field of view and distance of the ultrasonic radar.

[0090] Construct a mixed-integer programming model, including:

[0091] Four binary decision variables were defined to determine the placement of the four types of sensors:

[0092]

[0093] If a solid-state LiDAR is deployed at location s, then α s =1, otherwise α s =0;

[0094]

[0095] If a camera is placed at position c, then β c =1, otherwise β c=0;

[0096]

[0097] If a millimeter-wave radar is deployed at location m, then χ m =1, otherwise χ m =0;

[0098]

[0099] If an ultrasonic radar is deployed at position u, then δ u =1, otherwise δ u =0;

[0100] Construct a multi-objective optimization function:

[0101] O=min(ψ1O1+ψ2O2+ψ3O3) (5)

[0102] Where O1, O2, and O3 represent cost, coverage, and redundancy, respectively, and ψ1, ψ2, and ψ3 represent the weights of the three sub-objectives;

[0103] The cost is determined by the price of the sensor, as defined below:

[0104]

[0105] in and ω1, ω2, ω3, and ω4 represent the total number of solid-state lidar, cameras, millimeter-wave radar, and ultrasonic radar, respectively, while ω4 represents the price of the four types of sensors.

[0106] The perception area of ​​the autonomous vehicle to be configured is divided into a core perception area and an extended perception area. The coverage rate is defined as the weighted sum of the coverage rates of the core perception area and the extended perception area, as follows:

[0107]

[0108] Where S v1 and S v2 σ1 and σ2 respectively represent the coverage area of ​​the effective core sensing area and the area of ​​the extended sensing area, and represent the importance of the two types of coverage rates;

[0109] N sum The effective coverage area of ​​each sensor is defined as follows:

[0110]

[0111] Redundancy is the ratio of the total sensing area to the redundant coverage area of ​​multiple source sensors, defined as follows:

[0112]

[0113] Where S core ∪S extend S represents the area of ​​the perception zone for an autonomous vehicle. g and S h S represents the coverage area of ​​the sensor g and h. g ∩S h Corresponding to the area of ​​the redundant region covered by the two sensors, This represents the sum of the effective redundant areas covered by all sensors.

[0114] The mixed-integer programming model is constructed, satisfying the following constraints:

[0115] S core ×γ1≤S v1 (9)

[0116] Where γ1∈[0,1], the effective coverage core area of ​​the multi-source sensor cannot be less than γ1 times the core sensing area;

[0117] S extend ×γ2≤S v2 (10)

[0118] Where γ2∈[0,1], the effective coverage extension area of ​​the multi-source sensor cannot be less than γ2 times the extended sensing area;

[0119]

[0120] Where γ∈[0,1], the redundancy of the sensing area cannot be less than γ times the total sensing area;

[0121]

[0122] Where N S The integer is N. Considering the high cost of solid-state lidar, its quantity will not exceed N. S indivual;

[0123]

[0124] in and The integer is positive, and considering the possible placement locations of millimeter-wave radars, its number is in Within the range;

[0125]

[0126] in and The number of cameras is a positive integer, taking into account the possible placement locations of the cameras. Within the range;

[0127]

[0128] in and The number of ultrasonic radars is a positive integer, taking into account the possible placement locations. Within the range;

[0129] α s ∈{0,1},β c ∈{0,1},χ m ∈{0,1},δ u ∈{0,1} (16)

[0130] Ensure that the above variables are binary 0-1 variables.

[0131] Solving the mixed integer programming model includes: setting the objective function and parameters in the constraints, solving the model based on the genetic simulated annealing algorithm and CPLEX, and outputting the optimal configuration scheme for multiple sensors.

[0132] Example 2

[0133] like Figure 1 As shown in Embodiment 2, this method provides a multi-sensor optimization configuration method for autonomous vehicles. Based on the given vehicle structure and dimensions, solid-state LiDAR, cameras, millimeter-wave radar, and ultrasonic radar types are selected. Parameters such as field of view and distance are determined. The perception area of ​​the autonomous vehicle is divided into a core perception area and an extended perception area, which are then discretized. Based on the functions and perception ranges of various sensors, feasible sensor placement areas are defined. Considering cost, coverage, and redundancy, a quantitative optimization configuration scheme for multi-source sensors is determined. A mixed-integer programming model is constructed and solved using the genetic simulated annealing algorithm and CPLEX. The sensor placement positions and quantities are optimized, improving coverage and redundancy while reducing costs, resulting in the optimal configuration scheme. The specific steps are as follows:

[0134] Step 1. Determine the dimensions of the autonomous vehicle and parameters such as the field of view and distance of the solid-state LiDAR, cameras, millimeter-wave radar, and ultrasonic radar based on the autonomous vehicle model. Figure 2 As shown;

[0135] The specific parameters are as follows:

[0136] Solid-state LiDAR: Horizontal field of view 120° (-60° to +60°), range 200 meters;

[0137] Camera: Horizontal field of view 60° (-30° to +30°), distance 150 meters;

[0138] Millimeter wave radar: horizontal field of view 150° (-75° to +75°), range 100 meters;

[0139] Ultrasonic radar: Horizontal field of view 120° (-60° to +60°), range 2.5 meters.

[0140] Step 2. Divide the perception area according to the vehicle model, dividing the autonomous vehicle's perception area into a core perception area and an extended perception area. The coverage and redundancy requirements of the core perception area are higher than those of the extended perception area, such as... Figure 3 As shown;

[0141] Autonomous vehicle perception area P Z =P1∪P2∪P3∪P4∪P5∪P6, where the core sensing region is P1~P4, and its area is S. core The extended sensing area is P5 to P6, and its area is S. extend .

[0142] Step 3. Based on the functions and sensing ranges of various sensors, design a sensing area discretization mechanism, such as... Figure 4 As shown, the perception area of ​​an autonomous vehicle is discretized into a square region of size l*l. To simplify the computational complexity caused by the irregularity of the sensor coverage area, an approximate calculation method is proposed. When the covered square region is larger than l... If the coverage area is l2, then it is recorded as l2; otherwise, it is recorded as 0.

[0143] A feasible sensor placement area division method suitable for this vehicle model is proposed, such as... Figure 5 As shown, for different types of sensors, considering the characteristics of various sensors, feasible arrangement areas are set, where s={1,2,…,S}, c={1,2,…,C}, m={1,2,…,M}, u={1,2,…,U} represent the feasible arrangement position variables of solid-state lidar, camera, millimeter-wave radar and ultrasonic radar respectively, and S, C, M and U are positive integers.

[0144] Step 4. Considering cost, coverage and redundancy, determine the quantitative optimization configuration scheme of multi-source sensors, construct a mixed integer programming model, and obtain the optimal number of sensors and installation locations;

[0145] First, four binary decision variables were defined to determine the placement of four sensors: solid-state LiDAR, camera, millimeter-wave radar, and ultrasonic radar.

[0146]

[0147] If a solid-state LiDAR is deployed at location s, then α s =1, otherwise α s =0.

[0148]

[0149] If a camera is placed at position c, then β c =1, otherwise β c =0.

[0150]

[0151] If a millimeter-wave radar is deployed at location m, then χ m =1, otherwise χ m =0.

[0152]

[0153] If an ultrasonic radar is deployed at position u, then δ u =1, otherwise δ u =0.

[0154] Considering cost, coverage, and redundancy, a quantitative optimization configuration scheme for multi-source sensors is determined, and a multi-objective optimization function is constructed:

[0155] O=min(ψ1O1+ψ2O2+ψ3O3) (5)

[0156] Where O1, O2, and O3 represent cost, coverage, and redundancy, respectively, and ψ1, ψ2, and ψ3 represent the weights of the three sub-objectives.

[0157] The cost is determined by the price of the sensor, as defined below:

[0158]

[0159] in and ω1, ω2, ω3, and ω4 represent the total number of solid-state LiDAR, cameras, millimeter-wave radar, and ultrasonic radar, respectively, while ω4 represents the price of the four types of sensors (solid-state LiDAR, cameras, millimeter-wave radar, and ultrasonic radar).

[0160] Coverage is defined as the weighted sum of the coverage of the core sensing area and the coverage of the extended sensing area, as follows:

[0161]

[0162] Where S v1 and S v2 σ1 and σ2 represent the coverage area of ​​the effective core sensing area and the area of ​​the extended sensing area, respectively. σ1 and σ2 represent the importance of the two types of coverage, respectively, with σ1 usually being greater than σ2.

[0163] N sum The effective coverage area of ​​each sensor is defined as follows:

[0164]

[0165] Redundancy is the ratio of the total sensing area to the redundant coverage area of ​​multiple source sensors, defined as follows:

[0166]

[0167] Where S core ∪S extend S represents the area of ​​the perception zone for an autonomous vehicle. g and S h S represents the coverage area of ​​the sensor g and h. g ∩S h The area corresponding to the redundant region covered by the two sensors, such as Figure 4 As shown, This represents the sum of the effective redundant areas covered by all sensors.

[0168] The following constraints need to be met:

[0169] S core ×γ1≤S v1 (9)

[0170] Where γ1∈[0,1], the effective coverage core area of ​​the multi-source sensor cannot be less than γ1 times the core sensing area.

[0171] S extend ×γ2≤S v2 (10)

[0172] Where γ2∈[0,1], the effective coverage extension area of ​​the multi-source sensor cannot be less than γ2 times the extended sensing area.

[0173]

[0174] Where γ∈[0,1], the redundancy of the sensing area cannot be less than γ times the total sensing area.

[0175]

[0176] Where N S The integer is N. Considering the high cost of solid-state lidar, its quantity will not exceed N. S indivual.

[0177]

[0178] in and The integer is positive, and considering the possible placement locations of millimeter-wave radars, its number is in Within the range.

[0179]

[0180] in and The number of cameras is a positive integer, taking into account the possible placement locations of the cameras. Within the range.

[0181]

[0182] in and The number of ultrasonic radars is a positive integer, taking into account the possible placement locations. Within the range.

[0183] α s ∈{0,1},β c ∈{0,1},χ m ∈{0,1},δ u ∈{0,1} (16)

[0184] Ensure that the above variables are binary 0-1 variables.

[0185] Step 5. Define the parameters in the objective function and constraints, solve the problem using the genetic simulated annealing algorithm and CPLEX, and output the optimal configuration scheme for multiple sensors, such as... Figure 6 As shown.

[0186] Example 3

[0187] Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium for storing computer instructions. When executed by a processor, the computer instructions implement a multi-sensor optimization configuration method for autonomous vehicles. The method includes:

[0188] Obtain structural and dimensional parameters of the autonomous vehicle to be configured;

[0189] Based on the model of the autonomous vehicle, determine the perception area of ​​the autonomous vehicle to be configured, including the core perception area and the extended perception area.

[0190] Based on the parameter information of the autonomous vehicle to be configured, combined with the perception area of ​​the autonomous vehicle, and considering the functions and performance parameters of different types of sensors, the feasible placement areas of various sensors are determined according to the perception area discretization mechanism.

[0191] A mixed-integer programming model is constructed with the goal of minimizing sensor deployment cost, maximizing coverage and redundancy area.

[0192] Solve the mixed integer programming model to obtain the optimal number of sensors and their installation locations.

[0193] Example 4

[0194] Embodiment 4 of the present invention provides a computer program (product), including a computer program that, when run on one or more processors, is used to implement a multi-sensor optimization configuration method for autonomous vehicles. The method includes:

[0195] Obtain structural and dimensional parameters of the autonomous vehicle to be configured;

[0196] Based on the model of the autonomous vehicle, determine the perception area of ​​the autonomous vehicle to be configured, including the core perception area and the extended perception area.

[0197] Based on the parameter information of the autonomous vehicle to be configured, combined with the perception area of ​​the autonomous vehicle, and considering the functions and performance parameters of different types of sensors, the feasible placement areas of various sensors are determined according to the perception area discretization mechanism.

[0198] A mixed-integer programming model is constructed with the goal of minimizing sensor deployment cost, maximizing coverage and redundancy area.

[0199] Solve the mixed integer programming model to obtain the optimal number of sensors and their installation locations.

[0200] Example 5

[0201] Embodiment 5 of the present invention provides an electronic device, including: a processor, a memory, and a computer program; wherein, the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device is running, the processor executes the computer program stored in the memory to cause the electronic device to execute instructions for implementing a multi-sensor optimization configuration method for autonomous vehicles, the method including:

[0202] Obtain structural and dimensional parameters of the autonomous vehicle to be configured;

[0203] Based on the model of the autonomous vehicle, determine the perception area of ​​the autonomous vehicle to be configured, including the core perception area and the extended perception area.

[0204] Based on the parameter information of the autonomous vehicle to be configured, combined with the perception area of ​​the autonomous vehicle, and considering the functions and performance parameters of different types of sensors, the feasible placement areas of various sensors are determined according to the perception area discretization mechanism.

[0205] A mixed-integer programming model is constructed with the goal of minimizing sensor deployment cost, maximizing coverage and redundancy area.

[0206] Solve the mixed integer programming model to obtain the optimal number of sensors and their installation locations.

[0207] In summary, the multi-sensor optimization configuration method and system for autonomous vehicles described in this embodiment of the invention selects solid-state LiDAR, cameras, millimeter-wave radar, and ultrasonic radar types based on the given vehicle structure and size, determines parameters such as field of view and distance, divides the autonomous vehicle's perception area into a core perception area and an extended perception area, and discretizes them. Based on the functions and perception ranges of various sensors, feasible sensor placement areas are set. Considering cost, coverage, and redundancy, a quantitative optimization configuration scheme for multi-source sensors is determined. A mixed-integer programming model is constructed, model parameters are set according to requirements, and the model is solved based on the genetic simulated annealing algorithm and CPLEX. The sensor placement positions and quantities are optimized, improving coverage and redundancy areas while reducing costs, thus obtaining the optimal configuration scheme.

[0208] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0209] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0210] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0211] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment, whereby a series of operational steps are performed to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0212] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that, based on the technical solutions disclosed in the present invention, various modifications or variations that can be made by those skilled in the art without creative effort should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing the configuration of multiple sensors in an autonomous vehicle, characterized in that, include: Obtain parameter information of the autonomous vehicle to be configured; Based on the vehicle model, the perception area of ​​the autonomous vehicle to be configured is determined, including the core perception area and the extended perception area; the coverage rate is defined as the weighted sum of the coverage rates of the core perception area and the extended perception area, as follows: ； in and These represent the coverage area of ​​the effective core sensing area and the area of ​​the extended sensing area, respectively. and These respectively indicate the importance of the two types of coverage; This represents the area of ​​the core sensing region. Indicates the area of ​​the extended sensing region; The effective coverage area of ​​each sensor is defined as follows: ； Redundancy is the ratio of the total sensing area to the redundant coverage area of ​​multiple source sensors, defined as follows: ； in This represents the area of ​​the perception zone for autonomous vehicles. and Representative sensor and Coverage area Corresponding to the area of ​​the redundant region covered by the two sensors, This represents the sum of the effective redundant areas covered by all sensors; Based on the parameter information of the autonomous vehicle to be configured, combined with the perception area of ​​the autonomous vehicle, and considering the parameters of different types of sensors that need to be deployed, the feasible deployment areas of various sensors are determined according to the perception area discretization mechanism. To minimize sensor deployment cost, maximize coverage, and maximize redundancy, a mixed-integer programming model is constructed. The construction of the mixed-integer programming model includes: Four binary decision variables were defined to determine the placement of the four types of sensors: ； If a solid-state lidar is deployed at location s, then ,otherwise ; ； If a camera is placed at position c, then ,otherwise ; ； If a millimeter-wave radar is deployed at location m, then ,otherwise ; ； If an ultrasonic radar is deployed at position u, then ,otherwise ; Construct a multi-objective optimization function: ;in , and These represent cost, coverage, and redundancy, respectively. , and This indicates the weights of the three sub-objectives; The cost is determined by the price of the sensor, as defined below: ； in , , and These represent the total number of solid-state lidar, cameras, millimeter-wave radar, and ultrasonic radar, respectively. , , and These represent the prices of solid-state lidar, cameras, millimeter-wave radar, and ultrasonic radar, respectively. Solve the mixed integer programming model to obtain the optimal number of sensors and their installation locations.

2. The method for optimizing the configuration of multiple sensors in an autonomous vehicle according to claim 1, characterized in that, The parameter information of the autonomous vehicle to be configured includes: the structural and dimensional parameters of the autonomous vehicle to be configured; the parameters of the sensors to be deployed include the horizontal field of view and distance parameters of the sensors.

3. The method for optimizing the configuration of multiple sensors in an autonomous vehicle according to claim 1, characterized in that, The mixed-integer programming model is constructed, satisfying the following constraints: ;in The effective coverage area of ​​multi-source sensors must not be smaller than the core sensing area. times; ;in The effective coverage area of ​​a multi-source sensor cannot be less than the extended sensing area. times; ;in The redundancy range of the sensing area is not less than the total sensing area. times; ;in The number of solid-state lidar units is a positive integer, and considering the high cost of solid-state lidar, its quantity does not exceed [a certain number]. indivual; ;in and The integer is positive, and considering the possible placement locations of millimeter-wave radars, its number is in Within the range; ;in and The number of cameras is a positive integer, taking into account the possible placement locations of the cameras. Within the range; ;in and The number of ultrasonic radars is a positive integer, taking into account the possible placement locations. Within the range; , , , ； Ensure that the above variables are binary 0-1 variables.

4. The method for optimizing the configuration of multiple sensors in an autonomous vehicle according to claim 3, characterized in that, Solving the mixed integer programming model includes: setting the objective function and parameters in the constraints, solving the model based on the genetic simulated annealing algorithm and CPLEX, and outputting the optimal configuration scheme for multiple sensors.

5. An autonomous vehicle multi-sensor optimization configuration system based on the method described in any one of claims 1-4, characterized in that, include: The acquisition module is used to acquire parameter information of the autonomous vehicle to be configured; The perception area segmentation module is used to determine the perception area of ​​the autonomous vehicle to be configured, including the core perception area and the extended perception area, based on the vehicle model. The feasible deployment area construction module is used to determine the feasible deployment areas of various sensors based on the parameter information of the autonomous vehicle to be configured, combined with the perception area of ​​the autonomous vehicle, considering the parameters of different types of sensors that need to be deployed, and according to the perception area discretization mechanism. The planning model building module is used to construct a mixed-integer programming model with the goal of minimizing sensor deployment cost and maximizing coverage and redundancy. The solution module is used to solve the mixed integer programming model to obtain the optimal number of sensors and their installation locations.

6. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement the multi-sensor optimization configuration method for autonomous vehicles as described in any one of claims 1-4.

7. A computer program product, characterized in that, Includes a computer program, which, when run on one or more processors, is used to implement the multi-sensor optimization configuration method for autonomous vehicles as described in any one of claims 1-4.

8. An electronic device, characterized in that, include: The device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to cause the electronic device to execute instructions that implement the multi-sensor optimization configuration method for autonomous vehicles as described in any one of claims 1-4.

Images