A general emergency door lock opening method and system applied to a new energy vehicle

Abstract

The application provides a general emergency door lock opening method and system applied to a new energy vehicle, and belongs to the technical field of new energy vehicle emergency safety. The method comprises the following steps: collecting multi-source signal data of the vehicle in real time; pre-processing the collected multi-source signal data to generate a standardized signal data set; inputting the standardized signal data set into an improved fuzzy neural network decision model, combining a scene adaptive threshold value output in real time by a dynamic threshold value self-adaptive adjustment model to judge whether a vehicle door unlocking condition is met; and when it is determined that the unlocking condition is met, sending an unlocking instruction to a door lock controller by a vehicle control module. The application realizes accurate identification and rapid response to multiple dangerous scenes such as collision, battery overheating and power failure, solves the problems of high misjudgment rate and poor adaptability of a traditional scheme, and has high reliability, strong universality and continuous evolution capability.

Description

Technical Field

[0001] This invention belongs to the field of emergency safety technology for new energy vehicles, and in particular relates to a universal emergency door lock opening method and system applicable to new energy vehicle models. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] With the global energy structure transformation and the electrification upgrade of the automotive industry, the market penetration rate of new energy vehicles continues to rise, and their advantages in energy conservation, emission reduction, and cost control have been widely recognized. However, the complexity of the technical architecture, the special nature of the battery system, and the diversity of the vehicle body design of new energy vehicles have also given rise to new vehicle safety challenges. Among them, the emergency door unlocking problem in emergency situations is particularly prominent and directly relates to the life safety of drivers and passengers.

[0004] In dangerous scenarios such as severe collisions and battery thermal runaway, multiple risk factors often prevent vehicle doors from opening. Firstly, severe collisions can cause structural deformation of the vehicle body and jamming of the door lock mechanism, making physical opening of the doors difficult. Secondly, and more easily overlooked, is the failure of the electronic lock system. The impact of a collision can cause the main power connector to detach and the electrical control unit to malfunction, rendering the electronic door locks, which rely on main power, unresponsive, even if the door hardware is intact. Furthermore, the lack of effective means for quick unlocking from the outside further complicates the escape of trapped occupants.

[0005] Currently, the industry's exploration of technologies related to emergency vehicle unlocking still has significant limitations. Many existing solutions focus on triggering unlocking in a single collision scenario, relying solely on limited parameters such as collision signals and vehicle speed signals, failing to adequately cover dangerous scenarios unique to new energy vehicles, such as abnormal battery temperature and mains power failure. Furthermore, most solutions do not address the power supply guarantee for electronic locks after mains power failure, resulting in unlocking failure when the core power supply is interrupted. In addition, existing solutions do not incorporate intelligent models for deep fusion analysis of multi-source signals, making it difficult to distinguish between interference signals during normal driving and genuine danger signals, easily leading to mis-locking or missed unlocking issues. Summary of the Invention

[0006] To overcome the shortcomings of the prior art, this invention provides a universal emergency door lock opening method and system applicable to new energy vehicles. Through the synergistic effect of an improved fuzzy neural network decision model and a dynamic threshold adaptive adjustment model, it achieves accurate identification and rapid unlocking of dangerous scenarios, while ensuring the reliability of power supply and the universality of the strategy.

[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: The first aspect of this invention provides a universal emergency door lock opening method applicable to new energy vehicle models; A universal emergency door lock opening method applicable to new energy vehicle models, comprising: Real-time acquisition of multi-source signal data from vehicles; The acquired multi-source signal data is preprocessed to generate a standardized signal dataset; The standardized signal dataset is input into the improved fuzzy neural network decision model, and the scene adaptation threshold output by the model in real time is adjusted by dynamic threshold adaptive adjustment to determine whether the door unlocking condition is met. When the unlocking conditions are met, the vehicle control module sends an unlocking command to the door lock controller.

[0008] As a further technical solution, the real-time acquisition of multi-source signal data from vehicles includes: The sensor array collects vehicle collision safety signals, airbag deployment signals, deceleration signals, braking signals, battery temperature signals, battery voltage signals, and door lock status signals in real time.

[0009] As a further technical solution, the acquired multi-source signal data is preprocessed, including: The Kalman filter algorithm is used to filter and denoise the acquired multi-source signal data, and the min-max normalization algorithm is used to map the amplitude of each signal to a specific range.

[0010] As a further technical solution, the improved fuzzy neural network decision model includes the following sequentially connected components: The input layer is used to receive standardized signals; The fuzzification layer uses a Gaussian membership function to fuzzify the input signal, dividing it into multiple fuzziness levels; The fuzzy rule layer activates corresponding decision rules based on a pre-defined rule base for dangerous scenarios. The hidden layer of the neural network is used to optimize the weights of the rules in the fuzzy rule layer; The output layer outputs a value representing the confidence level of unlocking.

[0011] As a further technical solution, the dynamic threshold adaptive adjustment model takes the vehicle's real-time speed, battery state of charge, and road conditions as inputs, and dynamically adjusts the deceleration threshold and battery temperature threshold required for unlocking judgment through a support vector machine algorithm.

[0012] As a further technical solution, the door lock controller is powered by a constant emergency power supply independent of the vehicle's main power supply to drive the door lock to unlock.

[0013] As a further technical solution, the method also includes: based on the feedback data after unlocking, performing real-time iterative optimization of the parameters of the improved fuzzy neural network decision model using a reinforcement learning algorithm.

[0014] A second aspect of the present invention provides a universal emergency door lock opening system applicable to new energy vehicle models.

[0015] A universal emergency door lock opening system for use in new energy vehicles, comprising: The signal acquisition module is configured to: acquire multi-source signal data of the vehicle in real time; The preprocessing module is configured to preprocess the acquired multi-source signal data to generate a standardized signal dataset. The decision module is configured to: input the standardized signal dataset into the improved fuzzy neural network decision model, and combine the dynamic threshold to adaptively adjust the scene adaptation threshold output by the model in real time to determine whether the door unlocking conditions are met. The door lock execution module is configured to send an unlocking command to the door lock controller when the unlocking conditions are met.

[0016] A third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of a universal emergency door lock opening method for new energy vehicles as described in the first aspect of the present invention.

[0017] The fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of a universal emergency door lock opening method for new energy vehicles as described in the first aspect of the present invention.

[0018] The above one or more technical solutions have the following beneficial effects: (1) This method integrates collision safety signals, airbag signals, vehicle deceleration signals, braking signals, battery temperature and voltage signals, and door lock status signals to construct a multi-dimensional signal perception system, covering typical dangerous scenarios including frontal collisions, side collisions, rollovers, battery thermal runaway, instantaneous power failure, and emergency braking. Through the comprehensive judgment of the intelligent model, it ensures that emergency unlocking can be triggered in a timely and accurate manner when various real dangers occur, greatly improving the escape opportunities and safety of drivers and passengers.

[0019] (2) This invention introduces an improved fuzzy neural network decision model, combined with deep fusion of multi-source signals and a dynamic rule base, to achieve intelligent identification of various dangerous scenarios such as collisions, battery anomalies, power supply failures, and sudden deceleration. This significantly improves the accuracy and reliability of decision-making and effectively avoids the misjudgment problem caused by single signal triggering or fixed threshold judgment in traditional solutions. At the same time, a dynamic threshold adaptive adjustment model is adopted to dynamically adjust the unlock trigger threshold based on the real-time vehicle status, so that the system can flexibly adapt to different vehicle models and different driving scenarios without the need for hardware reconstruction or software rewriting for specific vehicle models, which significantly reduces the system deployment and adaptation costs.

[0020] (3) The present invention designs an independent constant power supply emergency power supply system, which adopts lithium-based energy storage power supply, and can continuously supply power to the door lock controller and actuator when the vehicle's main power supply fails, fundamentally solving the hidden danger that traditional electronic locks cannot be unlocked in the event of power failure.

[0021] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0022] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0023] Figure 1 This is a flowchart of the method in the first embodiment.

[0024] Figure 2 This is a system structure diagram of the second embodiment. Detailed Implementation

[0025] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0026] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0027] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0028] Example 1 This embodiment discloses a universal emergency door lock opening method for new energy vehicles. It collects multi-source signal data of the vehicle in real time through a sensor array, uses an improved fuzzy neural network decision model, and combines the adaptive threshold output by the model with dynamic threshold to make a comprehensive judgment. It realizes accurate identification and rapid response to various dangerous scenarios such as collision, battery overheating, and power failure, and has high reliability, strong versatility and adaptive evolution capability.

[0029] Specifically, such as Figure 1 As shown, a universal emergency door lock opening method applicable to new energy vehicles includes: Step S1: Real-time acquisition of multi-source signal data from the vehicle.

[0030] By deploying sensor arrays at key locations throughout the vehicle, multi-source signal data is collected in real time with a 10ms cycle. Specifically: collision sensors are installed at the front, sides, and rear of the vehicle to collect collision safety signals; an airbag trigger sensor is integrated into the airbag control module to obtain airbag deployment signals; a high-precision acceleration sensor is installed at the vehicle's center of gravity to collect longitudinal and lateral deceleration signals; a travel sensor is installed at the brake pedal or the brake signal is read from the vehicle's CAN bus; at least two temperature sensors are placed at key temperature measurement points inside the power battery pack to monitor battery temperature signals in real time; the total battery voltage signal is directly read through the battery management system; and microswitches or Hall sensors are embedded in the door lock mechanisms of the four doors to collect door lock / unlock status signals.

[0031] All sensors transmit the raw signals they collect to the central control module in real time via CAN bus or dedicated wiring harness, forming a multi-dimensional real-time data stream covering vehicle dynamics, battery status, and door lock status, providing a complete and high-frequency input foundation for subsequent intelligent decision-making.

[0032] Step S2: Preprocess the collected multi-source signal data to generate a standardized signal dataset.

[0033] The acquired multi-source signal data is first processed using the Kalman filter algorithm to effectively filter out high-frequency interference components caused by road bumps, electromagnetic interference, or sensor noise, while retaining the signal's abrupt change characteristics in dangerous scenarios to ensure the signal's authenticity and stability.

[0034] Subsequently, the filtered signals are normalized using a min-max normalization algorithm. Based on the historical statistical extreme values ​​of each signal type, the amplitudes of each signal are linearly mapped to a unified [0,1] interval. This process eliminates the influence of differences in dimensions and magnitudes between different sensors on the model input, forming a standard signal vector with a consistent numerical range. Finally, a signal dataset containing seven standardized feature values ​​is generated every 10ms, providing a structurally sound and feature-clear input for the subsequent intelligent decision-making model.

[0035] Step S3: Input the standardized signal dataset into the improved fuzzy neural network decision model, and combine the scene adaptation threshold output by the model in real time with the dynamic threshold adaptive adjustment to determine whether the door unlocking conditions are met.

[0036] Among them, the improved fuzzy neural network decision model is a feedforward inference system that deeply integrates fuzzy logic and artificial neural networks. Its data processing flow unfolds layer by layer from bottom to top, including: Input Layer: This layer serves as the data interface, receiving the preprocessed, standardized signal vector. Specifically, a vector containing seven standardized feature values ​​is simultaneously input to the corresponding input neurons in this layer. These neurons do not perform computations; they are only responsible for passing the data to the next layer.

[0037] Fuzzification Layer: This layer is the starting point for fuzzy logic processing. Its core function is to transform precise numerical inputs into fuzzy linguistic variables to simulate human qualitative judgments such as "mild," "severe," and "excessive." Each input neuron corresponds to a signal dimension and is connected to multiple fuzzy neurons in this layer. Specifically, each input signal is processed by three Gaussian membership functions to calculate its membership degree μ to the three fuzzy levels of "safe," "warning," and "danger," as shown below:

[0038] in, Let be the membership degree of the i-th signal belonging to the j-th fuzzy level. ; These correspond to "safety", "warning", and "danger" respectively; The center value of the j-th ambiguity level of the i-th signal; Let be the standard deviation of the j-th ambiguity level of the i-th signal.

[0039] Each input signal is processed by three Gaussian membership functions to obtain three membership values, which are used as input data for the fuzzy rule layer, thereby realizing the mapping from precise values ​​to fuzzy concepts.

[0040] Fuzzy Rule Layer: This layer encapsulates the system's domain knowledge and pre-defines a rule base containing 128 fuzzy rules. Based on the pre-defined rule base of 128 dangerous scenarios, rule matching is performed on the membership values ​​output by the fuzzification layer. The reasoning strength of each rule is calculated through product reasoning, realizing the logical association and fusion of multi-source signals.

[0041] The core formula is shown below:

[0042] in, Let be the inference strength of the k-th fuzzy rule; This is the fuzzy level index corresponding to the i-th signal in the k-th rule. For example, the rule "collision + battery overheating" corresponds to... (Collision hazard) (Temperature Hazard), other signals (Warning).

[0043] The 128 rules correspond to signal combination patterns in different typical dangerous scenarios. Each rule fuses the membership values ​​of multiple signals through a product operation, outputting a single inference strength. Ultimately, a 128-dimensional inference intensity vector is generated, reflecting the degree of matching between the current signal state and various dangerous scenarios.

[0044] The hidden layer of the neural network consists of 32 neurons and employs non-linear activation functions such as ReLU. It optimizes rule weights through backpropagation and performs non-linear feature fusion on the inference strength output of the fuzzy rule layer, improving the model's adaptability to complex scenarios. The input and output of the hidden layer are as follows:

[0045]

[0046] in, This is the input value of the m-th hidden layer neuron; The connection weights from the k-th rule to the m-th hidden layer neuron; This is the bias term for the m-th hidden layer neuron; This represents the output value of the m-th hidden layer neuron.

[0047] This process is equivalent to automatically learning and optimizing the relative importance (weights) of each fuzzy rule in the final decision, solving the drawback of traditional fuzzy systems where rule weights rely on expert experience and are difficult to adjust. The output of the hidden layer is a high-order feature representation after weighted synthesis.

[0048] Output layer: This layer receives the output of the hidden layer and is usually composed of a single neuron. It uses the sigmoid activation function to map the high-dimensional features of the hidden layer into a scalar value in the interval (0,1), which is the unlock confidence. This confidence quantitatively represents the probability that the scene indicated by the current combination of multi-source signals belongs to a real dangerous scene and therefore requires emergency unlocking, as shown below:

[0049]

[0050] in, The input values ​​are the values ​​for the output layer neurons; The connection weights from the m-th hidden layer neuron to the output layer; For the bias terms of the output layer neurons; To unlock confidence. When When the scenario is deemed a real danger, an emergency unlocking command is triggered; when If the signal is detected as interference, the unlocking operation will not be performed.

[0051] Furthermore, to improve decision-making accuracy, the model minimizes the confidence prediction error through backpropagation and updates it in real time. and With equal weighting parameters, and using mean squared error as the loss function, as follows:

[0052]

[0053] in, This represents the model loss value; This represents the number of training samples; Let be the prediction confidence level for the t-th sample; Let be the true confidence level of the t-th sample (1 for dangerous scenarios, 0 for normal scenarios); The updated weight values; The weight values ​​before the update; The learning rate controls the step size for weight updates; This is the partial derivative of the loss function with respect to the weights.

[0054] Through the weight iteration optimization driven by the above formula, the model can continuously learn the signal characteristics of different dangerous scenarios, gradually reduce prediction errors, and ensure dynamic improvement in decision accuracy.

[0055] Step S4: When it is determined that the unlocking conditions are met, the vehicle control module sends an unlocking command to the door lock controller.

[0056] Once the system determines that the unlocking conditions are met, the vehicle control module immediately sends a high-priority emergency unlocking command to the door lock controller via the vehicle's CAN bus. To ensure reliable execution of the command even in critical moments when the vehicle's main power supply may fail (e.g., due to a collision causing a power outage or battery failure), the door lock controller is powered entirely by a dedicated emergency power supply independent of the vehicle's high-voltage and 12V low-voltage systems. This emergency power supply utilizes a high-energy-density lithium-based energy storage unit, which is normally kept fully charged via the vehicle's charging system or low-voltage grid float charging. Upon receiving a digital or hardwired unlocking command from the vehicle control module, the door lock controller immediately drives its connected electromagnetic or motor-driven door lock actuator to overcome the mechanical lock and complete the unlocking action. The entire physical process, from command issuance to bolt return, is controlled within 10 milliseconds, ensuring that occupants have access to an unobstructed exit within the critical escape time.

[0057] Furthermore, in this embodiment, a continuously self-improving closed-loop optimization mechanism is constructed, including real-time iterative optimization of the parameters of the improved fuzzy neural network decision-making model based on feedback data after unlocking execution using a reinforcement learning algorithm. Specifically, each unlocking event, along with the multi-source signals before and after it, the decision confidence level, and the execution result, are stored as feedback data packets. Once the data accumulates to a set batch, the parameters of the fuzzy neural network decision-making model are iteratively optimized using a reinforcement learning algorithm, with the optimization goal of improving the correct decision rate and reducing the false / missed unlocking rate. The optimized model parameters are verified and then updated online, enabling the system to continuously learn and evolve from real-world scenarios.

[0058] Example 2 This embodiment discloses a universal emergency door lock opening system applicable to new energy vehicles; like Figure 2 As shown, a universal emergency door lock opening system for new energy vehicles includes: The signal acquisition module is configured to: acquire multi-source signal data of the vehicle in real time; The preprocessing module is configured to preprocess the acquired multi-source signal data to generate a standardized signal dataset. The decision module is configured to: input the standardized signal dataset into the improved fuzzy neural network decision model, and combine the dynamic threshold to adaptively adjust the scene adaptation threshold output by the model in real time to determine whether the door unlocking conditions are met. The door lock execution module is configured to send an unlocking command to the door lock controller when the unlocking conditions are met.

[0059] Example 3 The purpose of this embodiment is to provide a computer-readable storage medium.

[0060] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a universal emergency door lock opening method for new energy vehicles as described in Example 1.

[0061] Example 4 The purpose of this embodiment is to provide an electronic device.

[0062] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in a universal emergency door lock opening method for new energy vehicles as described in Embodiment 1.

[0063] The steps and methods involved in the apparatuses of Embodiments 2, 3, and 4 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0064] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0065] 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 various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A universal emergency door lock opening method applicable to new energy vehicle models, characterized in that, include: Real-time acquisition of multi-source signal data from vehicles; The acquired multi-source signal data is preprocessed to generate a standardized signal dataset; The standardized signal dataset is input into the improved fuzzy neural network decision model, and the scene adaptation threshold output by the model in real time is adjusted by dynamic threshold adaptive adjustment to determine whether the door unlocking condition is met. When the unlocking conditions are met, the vehicle control module sends an unlocking command to the door lock controller.

2. The universal emergency door lock opening method for new energy vehicles as described in claim 1, characterized in that, The real-time acquisition of multi-source signal data from vehicles includes: The sensor array collects vehicle collision safety signals, airbag deployment signals, deceleration signals, braking signals, battery temperature signals, battery voltage signals, and door lock status signals in real time.

3. The universal emergency door lock opening method for new energy vehicles as described in claim 1, characterized in that, Preprocessing of the acquired multi-source signal data includes: The Kalman filter algorithm is used to filter and denoise the acquired multi-source signal data, and the min-max normalization algorithm is used to map the amplitude of each signal to a specific range.

4. The universal emergency door lock opening method for new energy vehicles as described in claim 1, characterized in that, The improved fuzzy neural network decision model includes the following sequentially connected components: The input layer is used to receive standardized signals; The fuzzification layer uses a Gaussian membership function to fuzzify the input signal, dividing it into multiple fuzziness levels; The fuzzy rule layer activates corresponding decision rules based on a pre-defined rule base for dangerous scenarios. The hidden layer of the neural network is used to optimize the weights of the rules in the fuzzy rule layer; The output layer outputs a value representing the confidence level of unlocking.

5. The universal emergency door lock opening method for new energy vehicles as described in claim 1, characterized in that, The dynamic threshold adaptive adjustment model takes the vehicle's real-time speed, battery state of charge, and road conditions as inputs, and dynamically adjusts the deceleration threshold and battery temperature threshold required for unlocking judgment through a support vector machine algorithm.

6. The universal emergency door lock opening method for new energy vehicles as described in claim 1, characterized in that, The door lock controller is powered by a constant emergency power supply independent of the vehicle's main power supply, which drives the door lock to unlock.

7. The universal emergency door lock opening method for new energy vehicles as described in claim 1, characterized in that, The method further includes: based on the feedback data after unlocking, performing real-time iterative optimization of the parameters of the improved fuzzy neural network decision model using a reinforcement learning algorithm.

8. A universal emergency door lock opening system for new energy vehicles, characterized in that, include: The signal acquisition module is configured to: acquire multi-source signal data of the vehicle in real time; The preprocessing module is configured to preprocess the acquired multi-source signal data to generate a standardized signal dataset. The decision module is configured to: input the standardized signal dataset into the improved fuzzy neural network decision model, and combine the dynamic threshold to adaptively adjust the scene adaptation threshold output by the model in real time to determine whether the door unlocking conditions are met. The door lock execution module is configured to send an unlocking command to the door lock controller when the unlocking conditions are met.

9. A computer-readable storage medium having a program stored thereon, characterized in that, When executed by the processor, the program implements the steps of the universal emergency door lock opening method for new energy vehicles as described in any one of claims 1-7.

10. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the universal emergency door lock opening method for new energy vehicles as described in any one of claims 1-7.

Images