Battery car rental system capable of realizing identity authentication and remote management

Abstract

The application discloses a battery car leasing system capable of realizing identity authentication and remote management and control, and relates to the technical field of battery car leasing. The system comprises a cloud platform, a user terminal and a vehicle-mounted terminal. The vehicle-mounted terminal comprises a multi-modal acquisition module, a fusion authentication module and an accident sensing and response module. The multi-modal acquisition module acquires fingerprint, palm vein, facial feature fragments, riding posture data and PPG physiological signals. The fusion authentication module performs probability fusion on time-synchronized multi-source feature fragments, and outputs a continuous identity confidence. When the identity confidence is lower than a threshold, gradual driving right recovery is performed. The accident sensing and response module performs time alignment on heart rate variability mutation, grip attenuation and impact characteristics, calculates a response level index and determines an accident level, and generates a local active guardian instruction. The application realizes continuous identity authentication in the whole riding process and accurate accident grading response based on physiological and physical joint modeling, and improves the safety and reliability of the system.

Description

Technical Field

[0001] This invention relates to the field of electric bicycle rental technology, specifically to an electric bicycle rental system that enables identity verification and remote control. Background Technology

[0002] Existing electric bicycle rental systems typically consist of a cloud platform, user terminals, and onboard terminals installed on the electric bicycles, enabling user authentication and remote vehicle control. Regarding authentication, existing systems generally employ a static, one-time verification method. This involves comparing a single biometric feature, such as a fingerprint or facial image, collected by the user terminal when unlocking the vehicle. Once verification is successful, full driving privileges are granted. For remote control, existing systems primarily rely on onboard positioning modules and cellular network communication. The cloud platform issues remote control commands, such as speed limits or locking, based on the vehicle's location information.

[0003] However, the aforementioned existing technologies have the following shortcomings in practical applications:

[0004] 1. Static, one-time authentication methods cannot guarantee the consistency of identity throughout the entire riding process. After unlocking, users can hand over the vehicle to others, posing a security risk of identity theft. Some systems attempt to continuously collect biometric features such as facial features during riding, but due to the complex riding environment such as bumps and changes in lighting, the success rate of recognizing a single biometric feature is low, making it difficult to achieve reliable and continuous identity authentication.

[0005] 2. Existing accident detection mechanisms rely solely on physical signals such as acceleration sensors to determine whether a vehicle has fallen, failing to distinguish between a vehicle falling and actual rider injuries, resulting in a high false alarm rate. Furthermore, they lack the ability to perceive the rider's physiological state after an accident, making it difficult to assess the severity of the accident and providing effective injury prediction information for rescue efforts.

[0006] 3. The existing system relies heavily on cellular networks for identity verification and remote control. In areas with no signal, such as underground parking garages and remote scenic spots, users cannot complete normal car rental and return operations, which severely restricts the availability of the system and the user experience.

[0007] 4. Existing remote control strategies are relatively crude, usually implemented by directly cutting off power or locking the vehicle, lacking consideration for the traffic environment and dynamic right-of-way of the vehicle. Such sudden power interruption may cause secondary safety accidents. Summary of the Invention

[0008] (a) Technical problems to be solved

[0009] To address the shortcomings of existing technologies, this invention provides an electric vehicle rental system that enables identity verification and remote management, thus solving the problems mentioned in the background section.

[0010] (II) Technical Solution

[0011] To achieve the above objectives, the present invention provides the following technical solution: an electric bicycle rental system capable of identity verification and remote control, comprising:

[0012] The multimodal acquisition module is used to collect the cyclist's biometric signals, physiological signals, and vehicle motion status signals;

[0013] A short-range communication module is used to directly interact with the user terminal and exchange identity credentials in an offline state;

[0014] The fusion authentication module is connected to the multimodal acquisition module and is used to perform temporal correlation and probability fusion on the multi-source feature fragments output by the multimodal acquisition module to output a continuous identity confidence assessment result.

[0015] The accident perception and response module is connected to the multimodal acquisition module and is used to perform physiological and physical joint modeling based on physiological signals and vehicle motion state signals to determine whether an accident has occurred and the accident level, and generate corresponding local active protection commands.

[0016] The active execution module, connected to the fusion authentication module and the accident perception and response module, is used to respond to the identity confidence assessment result to perform a progressive revocation of driving rights operation, and to respond to the local active protection command to perform sound and light multimodal distress call and on-site protection actions.

[0017] Furthermore, the multimodal acquisition module includes: a biometric sensor mounted on the handlebars for acquiring fingerprint feature segments and palm vein feature segments; a camera mounted on the instrument panel area for acquiring facial feature segments; a PPG sensor integrated on the handlebars for acquiring photoplethysmography signals containing blood oxygen saturation and heart rate variability information; and an inertial measurement unit (IMU) for acquiring vehicle triaxial acceleration, angular velocity, and rider posture data.

[0018] Furthermore, the fusion authentication module is specifically configured as follows:

[0019] During cycling, the system periodically acquires a first feature segment output by the biometric sensor, a second feature segment output by the camera, and a cycling posture feature segment output by the IMU. After synchronizing the first feature segment, the second feature segment, and the cycling posture feature segment in time, these are input into a preset probability fusion model. The probability fusion model calculates the confidence probability value that the current cyclist is the same authorized user according to the following formula. :

[0020] ;

[0021] In the formula, The score for matching the first feature segment. The score for matching the second feature segment. Scoring is given for matching cycling posture feature segments. The weighting coefficient; when the confidence probability value When the value is below a preset threshold, a driving rights revocation trigger signal is output to the active execution module.

[0022] Furthermore, the fusion authentication module is also configured to: use the photoplethysmography signal output by the PPG sensor as a liveness detection auxiliary factor, and input it together with the first feature segment, the second feature segment and the riding posture feature segment into the probabilistic fusion model to improve the accuracy of identity fraud identification.

[0023] Furthermore, the accident perception and response module is specifically configured as follows:

[0024] The heart rate variability abrupt change feature output by the PPG sensor, the grip force decay curve feature output by the grip force sensor preset on the handlebars, and the impact feature output by the IMU are obtained; the heart rate variability abrupt change feature, the grip force decay curve feature, and the impact feature are time-aligned; based on the time alignment result, the response level index is calculated according to the following accident level determination formula. :

[0025] ;

[0026] In the formula, This represents the amplitude of heart rate variability mutations. Grip strength decay rate To achieve peak impact acceleration, To preset the impact threshold, The weighting coefficients are based on the response level index. The value range is used to distinguish between shock response, impact response and physiological collapse response, and output a judgment result that includes at least Level 1, Level 2 and Level 3 events; based on the judgment result, a local active protection command corresponding to the accident level is generated, and the parsed vital sign data contained in the command corresponding to the Level 3 event is converted into event status summary information and pushed to the emergency center information platform through the cellular network.

[0027] Furthermore, the short-range communication module is at least one of an NFC module and a Bluetooth module. The short-range communication module is configured to: in areas without cellular network coverage, perform two-way challenge-response authentication with the user terminal based on a valid digital certificate pre-stored in the user terminal; after successful authentication, generate an offline transaction voucher containing user identifier, rental and return time, and vehicle usage information and store it locally; when network recovery is detected or a communication connection is established with other vehicle terminals, upload the offline transaction voucher to the cloud platform for delayed settlement.

[0028] Furthermore, the vehicle-mounted terminal also includes a high-precision positioning module and a V2X communication module; the cloud platform is also used to generate a digital layer containing dynamic right-of-way rules and send it to the vehicle-mounted terminal; the vehicle-mounted terminal also includes a safety control module, which is configured to: when it is determined based on the high-precision positioning module and the digital layer that the current vehicle is about to enter a violation area or is deviating from the permitted driving area, output a flexible intervention command to the active execution module; the active execution module responds to the flexible intervention command and performs a stepped power limitation operation, which includes gradually reducing the motor output power, gradually limiting the maximum vehicle speed, and triggering directional voice prompts.

[0029] Furthermore, the local proactive protection commands generated by the accident perception and response module include:

[0030] Control the vehicle lights to flash a distress signal in a preset distress Morse code pattern;

[0031] The acoustic distress device is controlled to emit intermittent high-frequency location tones as acoustic guidance commands.

[0032] The V2X communication module broadcasts an early warning message containing the accident location information to surrounding vehicles.

[0033] (III) Beneficial Effects

[0034] 1. This invention periodically acquires fingerprint or palm vein feature fragments, facial feature fragments, and IMU-based cycling posture feature fragments through a multimodal acquisition module. A fusion authentication module then performs time synchronization and probabilistic fusion of these multi-source feature fragments, continuously outputting identity confidence assessment results. Even if a single sensor can only acquire incomplete feature fragments due to factors such as bumps, lighting changes, or occlusion, the probabilistic fusion of multi-source fragments can still maintain high authentication reliability.

[0035] 2. This invention, through an accident perception and response module, aligns the PPG heart rate variability mutation characteristics, grip strength decay curve characteristics, and IMU impact characteristics at the millisecond level. Based on physiological and physical joint modeling, it calculates a response level index, accurately distinguishing between shock reactions, impact reactions, and physiological collapse reactions, and outputting multi-level accident level judgments. According to the accident level, the system automatically executes differentiated local proactive protection commands, while simultaneously encapsulating the parsed vital sign data into an event status summary and pushing it to the emergency center, thus gaining crucial time for pre-hospital emergency care.

[0036] 3. This invention uses a short-range communication module to perform two-way challenge-response authentication with the user terminal based on a pre-stored digital certificate in offline mode, generating an offline transaction certificate and storing it locally; when the network is restored or communication is relayed through other vehicles, the offline transaction certificate is uploaded to the cloud platform to complete the delayed settlement, ensuring a normal rental and return experience in all scenarios.

[0037] 4. This invention combines a safety management module with high-precision positioning and a dynamic right-of-way digital layer. When a vehicle is about to enter a violation area, it outputs a flexible intervention command to the active execution module. This command performs a step-by-step power restriction operation, gradually reducing the motor output power, limiting the maximum speed, and triggering directional voice prompts. This guides the rider to actively return to a safe area, avoiding the safety risks caused by sudden power interruption. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the overall system structure framework of the present invention;

[0039] Figure 2 This is a flowchart of the integrated authentication module in this invention;

[0040] Figure 3 This is a flowchart of the accident perception and response module in this invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Please see Figures 1 to 3 As shown, the embodiments of the present invention provide the following technical solutions:

[0043] This embodiment discloses an electric bicycle rental system capable of identity verification and remote control, including a cloud platform, a user terminal, and an on-board terminal installed inside the electric bicycle. The cloud platform is responsible for user management, vehicle dispatching, data storage, and remote command issuance. The user terminal is a smartphone or other smart device with a rental application installed, used for data interaction with the cloud platform and the on-board terminal. The on-board terminal specifically includes a multimodal acquisition module, a short-range communication module, a fusion authentication module, an accident perception and response module, an active execution module, a high-precision positioning module, a V2X communication module, and a safety control module. Each module is described in detail below.

[0044] 1. Multimodal acquisition module, used to acquire the cyclist's biometric signals, physiological signals, and vehicle motion status signals.

[0045] The multimodal acquisition module includes a biometric sensor mounted on the handlebars for collecting fingerprint and palm vein feature fragments. In actual deployment, the biometric sensor can be an optical or capacitive fingerprint sensor, as well as a near-infrared palm vein sensor, both of which can be integrated into the same sensing module in the handlebar grip area. When the rider holds the handlebars normally, the palm and fingers naturally adhere to the sensor surface, and the system can complete the collection of feature fragments without the user's awareness.

[0046] A camera located in the instrument panel area is used to capture facial feature fragments; this camera is preferably an infrared binocular camera, capable of operating normally in low-light environments such as at night or in tunnels. The camera is installed at an angle facing the cyclist's face during normal riding, and its field of view covers the possible range of head movement.

[0047] The PPG sensor, integrated into the handlebars, is used to collect photoplethysmography (PPG) signals containing information on blood oxygen saturation and heart rate variability. The PPG sensor works by emitting specific wavelengths of green or infrared light into the skin and receiving the light signals reflected or transmitted through the vascular bed. Because blood volume changes periodically with the heartbeat, the intensity of the reflected light also exhibits a corresponding pulsating component. By analyzing this pulsating waveform, physiological parameters such as heart rate, blood oxygen saturation, and heart rate variability can be extracted. The PPG sensor is preferably integrated into the inside of the handlebars, in a position that naturally contacts the rider's palm, ensuring stable physiological signals throughout the ride. It is worth noting that this system completes all biometric and physiological signal analysis within the onboard terminal, transmitting only anonymized status indicators or alarm information to the cloud, without uploading any original biometric data that could identify an individual, thus complying with relevant laws and regulations on personal information protection.

[0048] An inertial measurement unit (IMU) is used to collect data on the vehicle's three-axis acceleration, angular velocity, and rider posture. An IMU typically includes a three-axis accelerometer and a three-axis gyroscope, and is installed on the vehicle's mainboard. By analyzing and processing the acceleration and angular velocity signals output by the IMU, the vehicle's pitch and roll attitudes, as well as the rider's control actions, can be extracted. Examples include the magnitude of forward weight shift during emergency braking and the lean angle of the vehicle when cornering. These features constitute unique segments of the rider's habitual riding posture.

[0049] 2. A short-range communication module, used to directly interact with the user terminal and exchange identity credentials in offline mode.

[0050] The short-range communication module is at least one of an NFC module and a Bluetooth module. The NFC module is suitable for users to tap their mobile phones close to the vehicle terminal to perform a tap operation. The Bluetooth module preferably supports the BLE low power protocol, with an effective communication distance of up to tens of meters, and is suitable for automatically completing identity pre-authentication when the user approaches the vehicle.

[0051] The short-range communication module is configured to perform a two-way challenge-response authentication with the user terminal in areas without cellular network coverage, such as underground parking lots, remote mountainous areas, or inside tunnels, based on a pre-stored valid digital certificate. This digital certificate, issued by the cloud platform during the user's most recent online login, includes the user's identity identifier, certificate validity period, and cloud signature. The specific process of the two-way challenge-response authentication is as follows: the vehicle terminal generates a random verification code and sends it to the user terminal; the user terminal signs the verification code using its private key or the key embedded in its certificate and returns it; the vehicle terminal verifies the signature using a pre-stored cloud public key or certificate chain; simultaneously, the user terminal also verifies the identity of the vehicle terminal to ensure the authenticity of both parties' identities. After successful authentication, the vehicle terminal generates an offline transaction voucher containing the user identifier, rental and return time, and vehicle usage information, and stores it locally. Vehicle usage information may include data such as riding mileage, duration, and battery consumption. When the vehicle-mounted terminal detects that the cellular network has been restored, or establishes a temporary communication connection with other passing electric vehicle-mounted terminals via the NFC module, the offline transaction voucher is uploaded to the cloud platform for delayed settlement via the relay link of the other vehicle-mounted terminals. This relay-style data upload mechanism between vehicles ensures that even if a vehicle is in a signal dead zone for an extended period, its offline transaction records can still be delivered to the cloud platform by leveraging the movement of other vehicles.

[0052] 3. A fusion authentication module, connected to the multimodal acquisition module, is used to perform temporal correlation and probabilistic fusion on the multi-source feature fragments output by the multimodal acquisition module, and output a continuous identity confidence assessment result. For details, please refer to [link to relevant documentation]. Figure 2 .

[0053] The fusion authentication module is specifically configured to: periodically acquire the first feature segment output by the biometric sensor, the second feature segment output by the camera, and the cycling posture feature segment output by the IMU during cycling; the acquisition period can be dynamically adjusted according to the cycling scenario, for example, a longer period can be used on straight road sections to reduce power consumption, and a shorter period can be used on bumpy road sections or when abnormal operation is detected to increase the authentication frequency.

[0054] After synchronizing the first feature segment, the second feature segment, and the cycling posture feature segment in time, they are input into a preset probabilistic fusion model. The specific implementation of time synchronization is as follows: each feature segment is assigned a unified system timestamp. The fusion authentication module pairs feature segments within the same acquisition window or those closest in time according to their timestamps, ensuring that the feature segments used for fusion calculation have a temporal correspondence. The probabilistic fusion model calculates the confidence probability value that the current cyclist is the same authorized user according to the following formula. :

[0055] ;

[0056] In the formula, The first feature segment matching score is obtained by performing local feature matching between the fingerprint or palm vein feature segment collected by the biometric sensor and the authorized user registration template. The value ranges from 0 to 1, and the higher the score, the higher the degree of matching. The second feature segment matching score is obtained by performing local feature matching between the facial feature segments captured by the camera and the authorized user's facial template, and the value ranges from 0 to 1. The matching score for riding posture feature segments is obtained by calculating the similarity between the control action features extracted by the current IMU and the riding posture model of the authorized user, with a value ranging from 0 to 1. The weighting coefficients are dynamically adjusted based on the signal quality and historical accuracy of each sensor in the current environment. For example, when the facial feature acquisition quality deteriorates in strong light, the weighting coefficients can be automatically reduced. And correspondingly improve and When wearing gloves prevents fingerprint features from being acquired, it can reduce... And improve and This ensures that the fusion authentication module can operate reliably under different working conditions. When the confidence probability value... When the value is below a preset threshold, a driving rights revocation trigger signal is output to the active execution module. The preset threshold can be set according to the system safety level requirements.

[0057] Furthermore, the fusion authentication module is configured to use the photoplethysmography (PPG) signal output by the PPG sensor as a liveness detection auxiliary factor, which is then input into the probabilistic fusion model along with the first feature segment, the second feature segment, and the cycling posture feature segment. The mechanism of the liveness detection auxiliary factor is as follows: the PPG signal reflects the pulsating characteristics of blood volume in a real human body, and its waveform exhibits micro-variability unique to living individuals, fundamentally different from the optical reflection characteristics of prosthetic materials such as photographs and silicone fingerprints; the fusion authentication module performs frequency and time domain analysis on the PPG signal, extracting liveness features such as heart rate variability. When the PPG signal is detected as lacking liveness features, regardless of the matching scores of other feature segments, it is directly determined to be a spoofing attack, thereby improving the accuracy of identity spoofing identification.

[0058] 4. The accident perception and response module, electrically connected to the multimodal acquisition module, is used to perform joint physiological and physical modeling based on physiological signals and vehicle motion state signals, determine whether an accident has occurred and its severity, and generate corresponding local active protection commands. For details, please refer to [link to relevant documentation]. Figure 3 .

[0059] The accident perception and response module is specifically configured to: acquire the heart rate variability mutation characteristics output by the PPG sensor, the grip force decay curve characteristics output by the grip force sensor preset on the handlebars, and the impact characteristics output by the IMU; wherein, the grip force sensor is preferably a thin-film piezoresistive sensor, attached below the grip area of ​​the handlebars, which can detect the pressure of the rider gripping the handlebars in real time. Under normal riding conditions, the rider's grip force on the handlebars is maintained within a relatively stable range; when an accident occurs that causes the rider to lose control or lose consciousness, the grip force curve shows a sharp decay until it reaches zero.

[0060] The heart rate variability mutation feature, the grip strength decay curve feature, and the impact feature are time-aligned; the time alignment accuracy can reach the millisecond level as needed, the purpose of which is to confirm whether the mutations of the three signals are caused by the same event. Specifically, the moment of the heart rate variability mutation, the moment when grip strength begins to decay, and the moment when the IMU detects the peak of impact acceleration are compared. If the three occur within a preset time window, they are determined to be caused by the same event. Based on the time alignment results, the response level index is calculated according to the following accident level determination formula. :

[0061] ;

[0062] In the formula, The amplitude of the heart rate variability mutation is expressed as the absolute change in heart rate variability before and after the accident, in milliseconds. The grip strength attenuation rate is expressed as the relative rate of change in grip strength before and after the accident, taking... This can make the growth of this item more gradual when the grip strength decay rate is large, thus avoiding the excessive contribution of a single indicator to the response level index. The peak impact acceleration is represented by the maximum value of the triaxial composite acceleration from the IMU at the moment of the accident. A preset impact threshold is set; the physical impact term only contributes positively to the response level index when the impact acceleration exceeds this threshold; otherwise, it does not. A value of 0 indicates that the physical impact is not significant. These are weighting coefficients, which can be obtained through machine learning training based on a large number of accident data samples.

[0063] According to the response level index The value range is used to distinguish between startle response, impact response, and physiological breakdown response, and the output should include at least the first-level event, second-level event, and third-level event determination results; for example, when When the value is less than the first preset threshold, it is judged as a startle response, corresponding to the first level event; when When the value is between the first and second thresholds, it is determined to be an impact response, corresponding to a level two event; when... When the value exceeds the second threshold, it is determined to be a physiological breakdown reaction, corresponding to a Level 3 event; those skilled in the art can set specific threshold boundary values ​​based on actual statistical data. Based on the determination result, a local active protection command corresponding to the incident level is generated, and the parsed vital sign data contained in the command corresponding to the Level 3 event is converted into event status summary information and pushed to the emergency center information platform via cellular network, providing decision-making reference for pre-hospital emergency care, but not constituting any form of medical diagnosis.

[0064] It should be noted that the event levels described in this invention are only internal classifications used to trigger differentiated vehicle safety response strategies and do not constitute a diagnosis or assessment of the rider's physical health or degree of injury; the physiological signals and their parsed data collected by the PPG sensor are only used to assist in determining whether an emergency rescue mechanism needs to be triggered and are not used as a basis for medical diagnosis. This system is not a medical device and does not provide any form of disease diagnosis or treatment function.

[0065] 5. An active execution module, electrically connected to the fusion authentication module and the accident perception and response module, is used to respond to the identity confidence assessment result to perform a progressive driving rights revocation operation, and to respond to the local active protection command to perform sound and light multimodal distress call and on-site protection actions.

[0066] When the fusion authentication module receives a driving rights revocation trigger signal, the active execution module performs a gradual driving rights revocation operation. This gradual operation is specifically manifested as follows: first, a warning is issued to the rider via a voice prompt module, such as "Authentication error, please confirm your identity"; if authentication is not restored within a certain time, the motor output power is gradually limited, for example, by reducing the output by a certain percentage every few seconds; finally, if authentication continues to fail, the maximum speed is limited to an extremely low value and the rider is guided to pull over. This gradual revocation method avoids the traffic accidents that may be caused by sudden power outages and vehicle locking in traditional solutions.

[0067] When responding to a local proactive protection command, the local proactive protection command generated by the accident perception and response module includes:

[0068] The vehicle lights are controlled to flash a distress signal in a preset Morse code pattern. The Morse code pattern is the internationally recognized three short, three long, and three short SOS signal. The control unit drives the front and rear lights of the vehicle to flash in this pattern in a cycle, which can be effectively identified at night or in low visibility conditions.

[0069] The acoustic distress signal device emits an intermittent high-frequency locating sound as an acoustic guidance command. The acoustic distress signal device can use a buzzer built into the vehicle or a separately installed piezoelectric generator to emit a high-frequency sound signal that repeats at a specific interval, which helps passersby or rescuers to quickly locate the accident vehicle and the injured along the direction of the sound.

[0070] The V2X communication module broadcasts an early warning message containing the accident location information to surrounding vehicles. The V2X communication module can use C-V2X or DSRC technology to send basic safety messages to other vehicles within the communication range via broadcast. The message contains the location coordinates of the accident vehicle and the accident level indicator. Surrounding vehicles that receive the message can display an accident warning on their in-vehicle screen to remind the driver to take precautions.

[0071] Furthermore, to achieve more refined proactive safety management, the high-precision positioning module supports RTK differential positioning, providing sub-meter level location information and distinguishing the specific lane the electric vehicle is currently in. The cloud platform is also used to generate a digital layer containing dynamic right-of-way rules and distribute it to the vehicle-mounted terminals. This dynamic right-of-way rule digital layer includes geofencing information such as non-motorized vehicle restricted areas, temporary construction zones, and bus stop areas on urban roads. The cloud platform can dynamically update this layer based on real-time announcements from municipal management departments and distribute it to each vehicle-mounted terminal via cellular network.

[0072] The safety control module is configured to: when it is determined, based on the high-precision positioning module and the digital layer, that the current vehicle is about to enter a violation area or is deviating from the permitted driving area, output a flexible intervention command to the active execution module; the active execution module responds to the flexible intervention command and performs a stepped power limitation operation, which includes gradually reducing the motor output power, gradually limiting the maximum vehicle speed, and triggering directional voice prompts, thereby achieving flexible control that balances management and safety.

[0073] It is worth noting that all safety control measures in this invention prioritize the safety of cyclists and public transportation. Any restrictions or interventions on driving rights are designed with safe execution conditions in mind to prevent sudden power interruptions during dangerous driving conditions. The biometric and physiological data collected in this invention are processed locally on the vehicle terminal and do not constitute improper collection or transmission of personal privacy. The accident emergency call and information push functions are solely for improving the efficiency of emergency rescue in road traffic accidents, and have a significant effect on enhancing public interest. In conclusion, the implementation of this invention does not violate national laws and does not harm the public interest.

[0074] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0075] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0076] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. An electric bicycle rental system capable of identity verification and remote control, characterized in that, It includes a cloud platform, a user terminal, and an on-board terminal installed inside the electric vehicle, wherein the on-board terminal includes: The multimodal acquisition module is used to collect the cyclist's biometric signals, physiological signals, and vehicle motion status signals; A short-range communication module is used to directly interact with the user terminal and exchange identity credentials in an offline state; The fusion authentication module is connected to the multimodal acquisition module and is used to perform temporal correlation and probability fusion on the multi-source feature fragments output by the multimodal acquisition module to output a continuous identity confidence assessment result. The accident perception and response module is connected to the multimodal acquisition module and is used to perform physiological and physical joint modeling based on physiological signals and vehicle motion state signals to determine whether an accident has occurred and the accident level, and generate corresponding local active protection commands. The active execution module, connected to the fusion authentication module and the accident perception and response module, is used to respond to the identity confidence assessment result to perform a progressive revocation of driving rights operation, and to respond to the local active protection command to perform sound and light multimodal distress call and on-site protection actions.

2. The electric bicycle rental system capable of identity verification and remote control according to claim 1, characterized in that, The multimodal acquisition module includes: a biometric sensor located on the handlebars for acquiring fingerprint and palm vein feature segments; a camera located in the instrument panel area for acquiring facial feature segments; a PPG sensor integrated on the handlebars for acquiring photoplethysmography (PPG) signals containing blood oxygen saturation and heart rate variability information; and an inertial measurement unit (IMU) for acquiring vehicle triaxial acceleration, angular velocity, and rider posture data.

3. The electric bicycle rental system capable of identity verification and remote control according to claim 2, characterized in that, The fusion authentication module is specifically configured as follows: During cycling, the system periodically acquires a first feature segment output by the biometric sensor, a second feature segment output by the camera, and a cycling posture feature segment output by the IMU. After synchronizing the first feature segment, the second feature segment, and the cycling posture feature segment in time, these are input into a preset probability fusion model. The probability fusion model calculates the confidence probability value that the current cyclist is the same authorized user according to the following formula. : ； In the formula, The score for matching the first feature segment. The score for matching the second feature segment. Scoring is given for matching cycling posture feature segments. The weighting coefficient; when the confidence probability value When the value is below a preset threshold, a driving rights revocation trigger signal is output to the active execution module.

4. The electric bicycle rental system capable of identity verification and remote control according to claim 3, characterized in that, The fusion authentication module is further configured to: use the photoplethysmography signal output by the PPG sensor as a liveness detection auxiliary factor, and input it together with the first feature segment, the second feature segment and the riding posture feature segment into the probabilistic fusion model to improve the accuracy of identity fraud identification.

5. The electric bicycle rental system capable of identity verification and remote control according to claim 2, characterized in that, The accident perception and response module is specifically configured as follows: The heart rate variability abrupt change feature output by the PPG sensor, the grip force decay curve feature output by the grip force sensor preset on the handlebars, and the impact feature output by the IMU are obtained; the heart rate variability abrupt change feature, the grip force decay curve feature, and the impact feature are time-aligned; based on the time alignment result, the response level index is calculated according to the following accident level determination formula. : ; In the formula, This represents the amplitude of heart rate variability mutations. Grip strength decay rate For peak impact acceleration, To preset the impact threshold, The weighting coefficients are based on the response level index. The value range is used to distinguish between shock response, impact response and physiological collapse response, and output a judgment result that includes at least Level 1, Level 2 and Level 3 events; based on the judgment result, a local active protection command corresponding to the accident level is generated, and the parsed vital sign data contained in the command corresponding to the Level 3 event is converted into event status summary information and pushed to the emergency center information platform through the cellular network.

6. The electric bicycle rental system capable of identity verification and remote control according to claim 1, characterized in that: The short-range communication module is at least one of an NFC module and a Bluetooth module. The short-range communication module is configured to: in areas without cellular network coverage, perform two-way challenge-response authentication with the user terminal based on a valid digital certificate pre-stored in the user terminal; after successful authentication, generate an offline transaction voucher containing user identifier, rental and return time, and vehicle usage information and store it locally; when network recovery is detected or a communication connection is established with other vehicle terminals, upload the offline transaction voucher to the cloud platform for delayed settlement.

7. The electric bicycle rental system capable of identity verification and remote control according to claim 1, characterized in that: The vehicle-mounted terminal also includes a high-precision positioning module and a V2X communication module; the cloud platform is also used to generate a digital layer containing dynamic right-of-way rules and send it to the vehicle-mounted terminal; the vehicle-mounted terminal also includes a safety control module, which is configured to: when it is determined based on the high-precision positioning module and the digital layer that the current vehicle is about to enter a violation area or is deviating from the permitted driving area, output a flexible intervention command to the active execution module; the active execution module responds to the flexible intervention command and performs a stepped power limitation operation when the safety execution conditions are met, the stepped power limitation operation includes gradually reducing the motor output power, gradually limiting the maximum vehicle speed, and triggering directional voice prompts.

8. The electric bicycle rental system capable of identity verification and remote control according to claim 5, characterized in that: The local proactive protection commands generated by the accident perception and response module include: Control the vehicle lights to flash a distress signal in a preset distress Morse code pattern; The acoustic distress device is controlled to emit intermittent high-frequency location tones as acoustic guidance commands. The V2X communication module broadcasts an early warning message containing the accident location information to surrounding vehicles.

Images