Method and system for predicting geometric conflicts between vehicle components

Abstract

The present disclosure provides a system (210) and a method (400) for predicting geometric conflicts between vehicle components. The method (400) includes collecting (410), by a controller (208) associated with the system (210), data associated with at least two components of the vehicle from a data system. The method (400) includes extracting (420), by the controller (208), one or more features of the at least two components from the collected data. The method (400) includes training (450), by the controller (208), a quantum circuit model based on the one or more features of the at least two components. The method (400) includes predicting (460), by the controller (208), the geometric conflicts between the at least two components via the quantum circuit model.

Description

Technical Field

[0001] This disclosure relates to the field of geometric conflict management. In particular, this disclosure provides a method and system for predicting geometric conflicts between vehicle components using quantum circuit models. Background Technology

[0002] Typically, vehicle components or parts undergo multiple design changes within a component system. During these design changes, the dimensions of these components may also change multiple times. These components may have a high tendency to collide or intersect with each other, a phenomenon known as geometric collision. The process of identifying valid geometric collisions between pairs of parts or components is called geometric collision management. In geometric collision management, a geometric collision can refer to the intersection between two or more parts or components. Intersections between two or more parts or components may occur during the design process due to non-compliance with part alignment or when parts are received from multiple suppliers.

[0003] Traditionally, multiple geometric conflicts have been reported in vehicles, and the classification of these conflicts may be performed or evaluated by the user, such as... Figure 1 As shown. Users may categorize geometric conflicts into relevant and irrelevant geometric conflicts based on their experience and historical data. Manual classification may not accurately assess or classify geometric conflicts. Users may need to open the data or a 3D geometric conflict matrix to provide a score each time. Therefore, it is necessary to identify geometric conflicts or interferences between two or more parts or components at an early stage to improve vehicle maturity and the quality of hardware prototypes in the early (digital) development phase.

[0004] Many technologies have been developed to eliminate the aforementioned problems. For example, patent document JP2009086864A describes a component interference inspection apparatus and method capable of performing interference inspections on all necessary components without exception. This component interference inspection apparatus has a database that categorizes the components constituting a vehicle into multiple evaluation units and stores component data associated with each component. An extraction unit (means) extracts component data associated with a first group of components and a second group of components composed of the same components from the database. The first and second component groups include all components belonging to the specified evaluation units. The interference inspection unit performs interference inspections by comparing the shortest distance between components with a predetermined reference distance, excluding combinations of the same components and combinations between components in the first and second component groups. A display unit displays the interference inspection results from the interference inspection unit.

[0005] Patent document US20080065251A1 describes a method and system for determining interference between a first physical part and at least a second part in the assembly of manufactured parts. The method includes: acquiring three-dimensional digital data representing the first physical part; using the digital data to form a virtual representation of the first physical part; placing the virtual representation of the first physical part and a virtual representation of the second part within a common reference frame; the second part corresponding to either a second physical part or a nominal part; and using the virtual representations of the first physical part and the second part to determine interference between the first physical part and the second part.

[0006] While the cited literature discloses various techniques for determining interference between two or more parts, these documents do not focus on predicting geometrical conflicts between two or more parts or components using quantum circuit models. Therefore, there remains room for solutions that can predict geometrical conflicts between two or more parts or components of a vehicle in advance.

[0007] The purpose of this disclosure

[0008] The overall objective of this disclosure is to provide an efficient and reliable system and method that eliminates the aforementioned limitations of existing systems and methods and anticipates geometric conflicts between two or more parts or components of a vehicle.

[0009] One object of this disclosure is to provide a system and method for extracting features of two or more parts or components from data collected from a data system.

[0010] Another object of this disclosure is to provide a system and method for training a quantum circuit model based on the features of the two or more parts or components.

[0011] Another object of this disclosure is to provide a system and method for predicting geometrical conflicts between the two or more parts or components via a quantum circuit model.

[0012] Another object of this disclosure is to provide a system and method that classifies geometric conflicts between two or more parts or components into relevant and unrelated geometric conflicts based on historical data and predefined configurations. Summary of the Invention

[0013] Various aspects of this disclosure relate to the field of geometric conflict management. In particular, this disclosure provides a method and system for predicting geometric conflicts between vehicle components using quantum circuit models.

[0014] One aspect of this disclosure relates to a method for predicting geometrical conflicts between vehicle components. The method includes: collecting data associated with at least two components of the vehicle from a data system by a controller associated with the system. The method includes: extracting one or more features of the at least two components from the collected data by the controller. The method includes: training a quantum circuit model by the controller based on the one or more features of the at least two components. The method includes: predicting the geometrical conflicts between the at least two components by the controller via the quantum circuit model.

[0015] In one aspect, the one or more features may include at least one of the following: the minimum distance between the at least two components, the Hausdorff distance between the at least two components, the coefficient of geometrical conflict between the at least two components, and the material of the at least two components.

[0016] In another aspect, the method may include: the controller determining the maximum thickness of the geometrical conflict between the at least two components based on the Hausdorff distance between the at least two components.

[0017] On the other hand, training the quantum circuit model by the controller may include: the controller merging the features of the at least two components with predefined coefficients; the controller identifying one or more outliers based on the merged features of the at least two components; and the controller scaling the collected data based on the one or more outliers.

[0018] In another aspect, the method may include: determining qubits by the controller based on the scaled data. The method may include: generating a quantum circuit in a grid lattice by the controller based on the qubits. The method may include: transforming the tensor corresponding to the scaled data into a quantum tensor by the controller by passing a tensor corresponding to the scaled data to the quantum circuit. The method may include: providing the quantum tensor as input to a neural network by the controller. The method may include: training the quantum circuit model by the controller based on the input.

[0019] In another aspect, the method may include: classifying the geometrical conflict between the at least two components into relevant geometrical conflicts and unrelated geometrical conflicts based on historical data and predefined configurations by the controller.

[0020] Another aspect of this disclosure relates to a system for predicting geometrical conflicts between vehicle components. The system includes: a controller associated with a processor, and a memory operatively coupled to the processor. The memory includes one or more instructions that, when executed, cause the controller to: collect data associated with at least two components of the vehicle from a data system; extract one or more features of the at least two components from the collected data; train a quantum circuit model based on the one or more features of the at least two components; and predict the geometrical conflict between the at least two components via the quantum circuit model.

[0021] In one aspect, the memory includes one or more instructions that, when executed, enable the controller to: determine the maximum thickness of the geometrical conflict between the at least two components based on the Hausdorff distance between the at least two components.

[0022] In one aspect, the controller can be configured to train the quantum circuit model by performing the following: merging the features of the at least two components with predefined coefficients; identifying one or more outliers based on the merged features of the at least two components; and scaling the collected data based on the one or more outliers.

[0023] On the other hand, the memory includes one or more instructions that, when executed, enable the controller to: determine qubits based on the scaled data; generate a quantum circuit in a grid lattice based on the qubits; transform the tensor corresponding to the scaled data into a quantum tensor by passing the tensor corresponding to the scaled data to the quantum circuit; provide the quantum tensor as input to a neural network; and train the quantum circuit model based on the input.

[0024] Various objects, features, aspects and advantages of the subject matter of this invention will become more apparent from the following detailed description of preferred embodiments and the accompanying drawings, in which the same reference numerals denote the same parts. Attached Figure Description

[0025] This disclosure includes accompanying drawings to provide a further understanding of the disclosure, and the drawings form part of this specification. The drawings illustrate exemplary embodiments of the disclosure and, together with this specification, serve to explain the principles of the disclosure.

[0026] Figure 1 This diagram illustrates an example flowchart for determining geometric conflicts between two or more components of a vehicle, based on existing technology.

[0027] Figure 2An example block diagram of a system for predicting geometrical conflicts between vehicle components is shown according to an embodiment of the present disclosure.

[0028] Figure 3 An example representation of an embodiment of the present disclosure is shown, illustrating a method for predicting geometrical conflicts between vehicle components.

[0029] Figure 4 A flowchart illustrating an example method for predicting geometrical conflicts between vehicle components, according to an embodiment of this disclosure, is shown.

[0030] Figure 5 illustrates an example computer system in which or by means of embodiments of this disclosure may be implemented.

[0031] The foregoing will become clearer from the following more detailed description of this disclosure. Detailed Implementation

[0032] The following is a detailed description of embodiments of the present disclosure depicted in the accompanying drawings. The details of these embodiments are sufficient to clearly convey the present disclosure. However, the amount of detail provided is not intended to limit the contemplation of the embodiments; rather, it is intended to cover all modifications, equivalents, and alternatives that fall within the spirit and scope of the present disclosure as defined by the appended claims.

[0033] The embodiments described herein relate to the field of geometric conflict management. In particular, this disclosure provides a method and system for predicting geometric conflicts between vehicle components using quantum circuit models. Various embodiments of this disclosure will be referenced. Figures 2 to 5 Please provide a detailed explanation.

[0034] refer to Figure 2 An example block diagram 200 of a proposed system (which may be interchangeably referred to herein as system 210) can be implemented in a vehicle for efficiently predicting geometrical conflicts between two or more components of the vehicle.

[0035] In one embodiment, system 210 may include one or more processors 202, which are implemented as one or more microprocessors, microcomputers, microcontrollers, edge or fog microcontrollers, digital signal processors, central processing units, logic circuits, and / or any device that processes data based on operating instructions. Among other functions, the one or more processors 202 may be configured to fetch and execute computer-readable instructions stored in memory 204 of system 210. Memory 204 may be configured to store one or more computer-readable instructions or routines in a non-transitory computer-readable storage medium, which may be fetched and executed to predict geometrical conflicts between two or more components or parts. Memory 204 may include any non-transitory storage device, including, for example, volatile memory (such as random access memory (RAM)) or non-volatile memory (such as erasable programmable read-only memory (EPROM), flash memory, etc.).

[0036] In one embodiment, system 210 may include interface 206. Interface 206 may include various interfaces, such as interfaces for data input / output devices (referred to as I / O devices), storage devices, etc. Interface 206 may facilitate communication of system 210. Interface 206 may also provide communication paths for one or more components of system 210. Examples of such components include, but are not limited to, controller 208 and database 222. Database 222 may include data generated or stored by functions implemented by processor 202, or controller 208, or any component of system 210.

[0037] According to one embodiment, controller 208 may include: a data collection engine 212, a feature extraction engine 214, a model training engine 216, a geometric conflict prediction engine 218, and other engines 220. These other engines 220 may implement functions that complement the applications / functions performed by controller 208. These other engines 220 may include one or more components selected from classification engines, detection engines, monitoring engines, etc.

[0038] In one embodiment, controller 208 may be associated with one or more processors 202, and memory 204 may be operatively coupled to one or more processors 202. Memory 204 may include one or more instructions that, when executed, cause controller 208 to: collect data associated with at least two components of the vehicle from a data system via data collection engine 212.

[0039] In one embodiment, controller 208 may extract one or more features of the at least two components from the collected data via feature extraction engine 214. The one or more features may include, but are not limited to: the minimum distance between the at least two components, the Hausdorff distance between the at least two components, the coefficient of geometrical conflict between the at least two components, and the material of the at least two components.

[0040] In one embodiment, if the minimum distance between the at least two components is positive, the controller 208 determines that there is no geometric conflict and the at least two components maintain a certain distance from each other. In one embodiment, if the minimum distance between the at least two components is negative, the controller 208 determines that there is a possibility of a geometric conflict and the at least two components intersect each other.

[0041] In one embodiment, controller 208 may determine the Hausdorff distance between the at least two components to determine the maximum thickness of the intersection between the at least two components. The Hausdorff distance between the at least two components may be a positive value.

[0042] In one embodiment, controller 208 may train a quantum circuit model based on one or more features of the at least two components via model training engine 216.

[0043] In another embodiment, controller 208 may combine one or more features of the at least two components with predefined coefficients (e.g., Trafo coefficients) via model training engine 216. In another embodiment, controller 208 may identify one or more outliers based on the one or more combined features of the at least two components. In another embodiment, controller 208 may scale the collected data based on the one or more outliers.

[0044] In one embodiment, controller 208 may determine qubits based on the scaled data via model training engine 216. In another embodiment, controller 208 may generate quantum circuits in a grid lattice based on the qubits. In yet another embodiment, controller 208 may transform the tensor corresponding to the scaled data into a quantum tensor by passing a tensor corresponding to the scaled data to the quantum circuit. In yet another embodiment, controller 208 may provide the quantum tensor as input to a neural network. In yet another embodiment, controller 208 may train the quantum circuit model based on the input via model training engine 216.

[0045] In one embodiment, controller 208 may use the quantum circuit model to predict geometric conflicts between the at least two components via a geometric conflict prediction engine 218. In another embodiment, controller 208 may classify geometric conflicts between the at least two components into relevant and irrelevant geometric conflicts based on historical data and predefined configurations via a classification engine (e.g., 220).

[0046] In one embodiment, controller 208 may determine the maximum thickness of the geometrical conflict between the at least two components based on the Hausdorff distance between them. The maximum thickness of the geometrical conflict between the at least two components can be determined to predict the geometrical conflict between them.

[0047] Figure 3 An example representation 300 illustrating an embodiment of the present disclosure of a method for predicting geometrical conflicts between vehicle components.

[0048] In classical computing, a bit may have only a single state: 0 or 1. In quantum computing, a qubit is the basic unit of information in a quantum state and can exist in a superposition of two states simultaneously. For a circuit composed of N qubits, there are 2^N possible combinations.

[0049] Computational platforms, such as quantum computers, can be used to train neural networks. Variation quantum circuits can be constructed or generated for quantum deep learning on noisy intermediate-scale quantum devices. A hybrid quantum-classical neural network architecture—where each neuron is a variation quantum circuit—can be used to perform geometric collision prediction for at least two parts of a vehicle. The performance of this hybrid quantum-classical neural network can be analyzed on a range of binary classification datasets, for example, using a simulated universal quantum computer.

[0050] refer to Figure 3In step 310, one or more features of the at least two components can be extracted from data collected from the data system. These features can be combined with predefined coefficients (i.e., Trafo coefficients). Outliers can be identified based on the combined features of the at least two components. The collected data can be scaled or preprocessed based on these one or more outliers.

[0051] In step 320, qubits can be determined based on the scaled data. Quantum circuits can be generated in a grid lattice based on the qubits. The tensor corresponding to the scaled data can be transformed into a quantum tensor by transferring it to the quantum circuit.

[0052] In step 330, the quantum tensor can be provided as input to the neural network through a classical layer to train the quantum circuit model based on the input. Therefore, the quantum circuit model can be a classical-quantum hybrid model. This classical-quantum hybrid model and the variable quantum circuit can be used for efficient and rapid model training. The output model can then be trained using the extracted features, enabling it to classify future geometric conflicts without any human intervention.

[0053] Figure 4 A flowchart is shown illustrating an example method 400 for predicting geometrical conflicts between vehicle components, according to an embodiment of the present disclosure.

[0054] refer to Figure 4 In step 410, method 400 may include: collecting data from a data system. The data collection may be performed by extracting desired attributes of components in tabular format from a pair table or lookup table.

[0055] In step 420, method 400 may include: extracting one or more features of the at least two components from the collected data. The one or more features may include, but are not limited to: the minimum distance between the at least two components, the Hausdorff distance between the at least two components, the coefficient of geometrical conflict between the at least two components, and the material of the at least two components.

[0056] In step 430, method 400 may include: data preparation. The data may be prepared by: merging one or more features of the at least two components with predefined coefficients; identifying outliers based on the merged features; and scaling and preprocessing the collected data based on the outliers.

[0057] In step 440, method 400 may include: initializing quantum machine learning (QML) using necessary qubits. The QML may be initialized based on model type, number of hidden layers, input tensor dimension, and optimizer type.

[0058] In step 450, method 400 may include: determining qubits based on the scaled data; and generating a quantum circuit in a grid lattice based on the qubits. Method 400 may include: transforming the tensor corresponding to the scaled data into a quantum tensor by passing a tensor corresponding to the scaled data to the quantum circuit. Method 400 may include: providing the quantum tensor as input to a neural network; and training the quantum circuit model based on the input to predict the geometric conflict between the at least two components.

[0059] In step 460, method 400 may include: predicting the geometric conflict between the at least two components. Method 400 may include: classifying the geometric conflict between the at least two components into relevant geometric conflicts and irrelevant geometric conflicts based on historical data and predefined configurations. In another embodiment, method 400 may include: determining the relevant and irrelevant geometric conflicts using the quantum circuit model.

[0060] Figure 5 illustrates an example computer system 500 in which or by means of embodiments of the present disclosure may be implemented.

[0061] refer to Figure 5 The computer system 500 includes: an external storage device 510, a bus 520, a main memory 530, a read-only memory 540, a mass storage device 550, a communication port 560, and a processor 570. Those skilled in the art will understand that the computer system 500 may include more than one processor 570 and communication port 560. The processor 570 may include various modules associated with embodiments of the present invention. The communication port 560 may be any of the following: an RS-232 port for use with a modem-based dial-up connection, a 10 / 100 Ethernet port, a gigabit or 10 gigabit port using copper or fiber optic cables, a serial port, a parallel port, or any other existing or future port. The communication port 560 may be selected based on the network, such as a local area network (LAN), a wide area network (WAN), or any other network connected to the computer system 500.

[0062] In one embodiment, memory 530 may be random access memory (RAM) or any other dynamic storage device known in the art. Read-only memory 540 may be any static storage device, such as, but not limited to, a programmable read-only memory (PROM) chip for storing static information (e.g., boot or BIOS instructions for processor 570). Mass storage device 560 may be any current or future mass storage solution that can be used to store information and / or instructions. Exemplary mass storage solutions include, but are not limited to, parallel advanced technology accessory (PATA) or serial advanced technology accessory (SATA) hard disk drives or solid-state drives (internal or external, such as having a universal serial bus (USB) and / or FireWire interface), one or more optical discs, and redundant array of independent disks (RAID) storage (e.g., disk arrays, such as SATA arrays).

[0063] In one embodiment, bus 520 can communicatively couple processor 570 with other memory, storage devices, and communication blocks. Bus 520 can be, for example, a Peripheral Component Interconnect (PCI) / PCI Expansion (PCI-X) bus, a Small Computer System Interface (SCSI) bus, a Universal Serial Bus (USB), etc., used to connect expansion cards, drives, and other subsystems, as well as other buses (such as the Front Side Bus (FSB), which is used to connect processor 570 to a software system).

[0064] In another embodiment, operator and management interfaces (e.g., a display, keyboard, and cursor control device) may also be coupled to bus 520 to support direct interaction between the operator and computer system 500. Other operator and management interfaces may also be provided via a network connection connected through communication port 560. External storage device 510 may be any type of external hard disk drive, floppy disk drive, read-only optical disc (CD-ROM), rewritable optical disc (CD-RW), or read-only digital video optical disc (DVD-ROM). The above components are intended only to illustrate various possible implementations. The above-described example computer system 500 should not, in any sense, limit the scope of this disclosure.

[0065] While various embodiments of the invention have been described for the foregoing, other and further embodiments of the invention may also be conceived without departing from the basic scope of the invention. The scope of the invention is defined by the appended claims. The invention is not limited to the described embodiments, versions, or examples, and the described embodiments, versions, or examples contained herein are intended only to enable those skilled in the art to make and use the invention in conjunction with the information and knowledge available to them.

[0066] Advantages of this disclosure

[0067] This disclosure provides a system and method for efficiently predicting geometrical conflicts between two or more parts or components of a vehicle.

[0068] This disclosure provides a system and method for extracting features of two or more parts or components from data collected from a data system.

[0069] This disclosure provides a system and method for training a quantum circuit model based on the features of the two or more parts or components.

[0070] This disclosure provides a system and method for predicting geometrical conflicts between the two or more parts or components via a quantum circuit model.

[0071] This disclosure provides a system and method for classifying geometric conflicts between two or more parts or components into relevant and unrelated geometric conflicts based on historical data and predefined configurations.

Claims

1. A method (400) for predicting geometrical conflicts between vehicle components, the method (400) comprising: The controller (208) associated with the system (210) collects (410) data associated with at least two components of the vehicle from the data system; The controller (208) extracts (420) one or more features of the at least two components from the collected data; The controller (208) trains (450) a quantum circuit model based on one or more features of the at least two components; as well as The controller (208) predicts (460) the geometric conflict between the at least two components via the quantum circuit model.

2. The method (400) of claim 1, wherein the one or more features include at least one of the following: minimum distance between the at least two components, Hausdorff distance between the at least two components, coefficient of the geometric conflict between the at least two components, and the material of the at least two components.

3. The method (400) as claimed in claim 2, comprising: The controller (208) determines the maximum thickness of the geometric conflict between the at least two components based on the Hausdorff distance between the at least two components.

4. The method of claim 1, wherein, Training (450) the quantum circuit model by the controller (208) includes: The controller (208) combines one or more features of the at least two components with predefined coefficients (430). The controller (208) identifies one or more outliers based on one or more merged features of the at least two components; and The controller (208) scales the collected data based on the one or more outliers.

5. The method (400) of claim 4, comprising: The controller (208) determines (440) qubits based on the scaled data; The controller (208) generates a quantum circuit in the grid lattice based on the qubit; The controller (208) transforms the tensor corresponding to the scaled data into a quantum tensor by passing the tensor corresponding to the scaled data to the quantum circuit; The controller (208) provides the quantum tensor as input to the neural network; as well as The controller (208) trains (450) the quantum circuit model based on the input.

6. The method (400) as claimed in claim 1, comprising: The controller (208) classifies the geometric conflicts between the at least two components into relevant geometric conflicts and unrelated geometric conflicts based on historical data and predefined configurations.

7. A system (210) for predicting geometrical conflicts between vehicle components, the system (210) comprising: A controller (208) associated with the processor (202); as well as A memory (204) operably coupled to the processor (202), wherein the memory (204) includes one or more instructions, which, when executed, cause the controller (208) to: From the data system, collect data associated with at least two components of the vehicle; From the collected data, extract one or more features of the at least two components; A quantum circuit model is trained based on one or more features of the at least two components; and The geometric conflict between the at least two components is predicted using the quantum circuit model.

8. The system (210) as claimed in claim 7, wherein, The memory (204) includes one or more instructions that, when executed, cause the controller (208) to: determine the maximum thickness of the geometric conflict between the at least two components based on the Hausdorff distance between the at least two components.

9. The system (210) as claimed in claim 7, wherein, The controller (208) is configured to train the quantum circuit model by performing the following: Combine one or more features of the at least two components with predefined coefficients; Based on the one or more merged features of the at least two components, identify one or more outliers; as well as The collected data is scaled based on the one or more outliers.

10. The system (210) as claimed in claim 9, wherein, The memory (204) includes one or more instructions that, when executed, cause the controller (208) to: Based on the scaled data, determine the qubits; Quantum circuits are generated in a grid lattice based on the aforementioned qubits; The tensor corresponding to the scaled data is transformed into a quantum tensor by passing it to the quantum circuit. The quantum tensor is provided as input to the neural network; as well as The quantum circuit model is trained based on the input.

Images