52 results about "Active safety systems" patented technology

The invention discloses an anti-rollover control method for a four-wheel steering vehicle based on deep reinforcement learning, and the method achieves the collaborative optimization of a four-wheel steering angle and a braking force through constructing a reinforcement learning frame of hybrid model driving and data driving. A hierarchical network structure is adopted to separate stability decision and execution control, and the generalization ability of the algorithm is improved in combination with virtual-real transfer learning. The anti-rollover control robustness under the complex working condition is remarkably improved, and the method can be applied to an active safety system of an intelligent driving vehicle.

The invention discloses a vehicle obstacle avoidance control method and device, and the method comprises the steps: predicting a vehicle driving track according to the environment sensing data of a vehicle and the vehicle driving state parameters under the condition that the vehicle driving state parameters meet a low-speed driving condition; respectively constructing vehicle contour envelopes for a plurality of vehicle track points in the vehicle driving track so as to generate corresponding vehicle contour envelope sequences; detecting a target vehicle contour envelope which is spatially overlapped with the obstacle for the first time from the vehicle contour envelope sequence, and determining obstacle collision time based on a target trajectory prediction point moment corresponding to the target polygon; and according to the obstacle collision time and the accelerator pedal state information of the vehicle, vehicle obstacle avoidance cooperative control operation is executed. Therefore, false alarm and false braking in the low-speed working condition are remarkably reduced, and the recognition sensitivity of the active safety system to the real collision risk is improved.

The invention belongs to the technical field of intelligent traffic and Internet of Vehicles, particularly relates to vehicle external environment detection equipment based on V2X, and aims to solve the problems that an existing scheme proposed in the background technology focuses on data sharing and lacks deep fusion and collaborative modeling capability of multi-source heterogeneous data. Comprising a V2X receiving and transmitting device supporting DSRC and C-V2X dual-mode communication protocols. By dynamically adjusting the weight of the sensor, the environment dependence of a single sensor is remarkably reduced, all-weather high-precision detection is realized, the problem of mismatching of multi-sensor data caused by clock skew or coordinate system difference is solved, a lane can be triggered in advance to keep assistance or the vehicle following distance can be adjusted, serial collision is effectively avoided, and the detection accuracy is improved. Potential cut-in risks can be predicted in advance, the response speed of an active safety system is remarkably improved, a path planning module generates a detouring track accordingly, the blind area defect of a traditional sensor is overcome, and a three-level early warning mechanism is triggered based on a fusion result of a D-S evidence theory and equipment.

The invention, which relates to the technical field of target detection, discloses a target detection method for reducing a vehicle collision risk, comprising: acquiring real-time image data of an environment in front of a vehicle; inputting the real-time image data into an optimized target detection neural network model, and identifying a pedestrian target in the image and position information of the pedestrian target; wherein the neural network model is based on a YOLO architecture, and optimization is carried out for a pedestrian detection task; a collision risk is calculated based on the identified pedestrian location information, and an active safety system of the vehicle is triggered when the risk exceeds a threshold. According to the scheme, the detection precision and real-time performance of the intelligent driving vehicle on pedestrians, especially small-scale pedestrians in a complex traffic environment can be remarkably improved, the collision risk caused by missing detection or false detection is effectively reduced, and the active safety performance of the vehicle is improved.

A system and method in which road markings, such as center-lane, shoulder markings and vertically oriented sign markings, are surface coatings configured to transmit machine-readable data to electronic equipment on self-driving or driving-assisted vehicles. In an embodiment, painted markings create a type of matrix code or barcode that can be read by optical detection systems in a vehicle. The data is used to inform active safety systems about road conditions, obstacles, grades, speed limits, turns, HOV, EV, or autonomous lane options, as well as other data, to safely conduct autonomous and semi-autonomous vehicles. In some embodiments markings are applied with coatings that are readable by electronic components and cameras onboard a vehicle, and may not be visible to humans.

The application discloses a vehicle obstacle avoidance control method and device, wherein the method comprises: in the case that a vehicle running state parameter meets a low-speed running condition, predicting a vehicle running track according to environmental perception data of the vehicle and the vehicle running state parameter; constructing a vehicle contour envelope for each vehicle track point in the vehicle running track to generate a corresponding vehicle contour envelope sequence; detecting a target vehicle contour envelope which first spatially overlaps with an obstacle from the vehicle contour envelope sequence, and determining an obstacle collision time based on a target track prediction point time corresponding to the target polygon; and performing a vehicle obstacle avoidance collaborative control operation according to the obstacle collision time and accelerator pedal state information of the vehicle. Thus, false alarms and false braking in a low-speed working condition are significantly reduced, and the identification sensitivity of an active safety system to a real collision risk is improved.

The invention discloses a method and device for actively preventing a vehicle from colliding with a person and the vehicle, and belongs to the technical field of vehicle active safety. The method comprises the steps that the real-time position of a vehicle to be driven is compared with a preset electronic fence database, and when it is judged that the vehicle enters a crowded key area, a protection system is forcibly activated; after the system is activated, a special surrounding person identification model is operated to enhance perception and accurately identify surrounding pedestrians, children, baby carriages, wheelchairs and other targets; synchronously fusing the target information and the state of the vehicle, and dynamically evaluating the collision risk level; and finally, according to the risk level and in combination with operation feedback of a driver, graded active control from acousto-optic warning, partial braking and full-force automatic braking to cooperative steering emergency risk avoiding is executed. Aiming at the defects that an existing active safety system is insufficient in protection for personnel outside a vehicle and can be manually closed, intelligent active protection which cannot be shielded in a key area and takes pedestrians as a core is achieved, the risk that the vehicle is used for malicious collision or is out of control to cause serious accidents is effectively reduced, and the system is easy to deploy based on an existing vehicle platform.

Provided herein is technology relating to automated driving, particularly, but not exclusively, to an enterprise active safety system (EASS) that is configured to improve safety or minimize losses from safety events. More specifically, the EASS is configured to provide system-based safety functions by integrating one or more of the vehicle components, road components, cloud components, pedestrian components, and map components. These safety functions include one or more of crash detection, crash prediction, crash warning, crash avoidance, crash impact reduction, and emergency response functions.

The application discloses an AFS, ARS and DYC cooperative control method of a distributed electric drive vehicle based on DDPG. The method utilizes a cooperative control method and a deep reinforcement learning method, designs and develops a vehicle stability control method based on a centroid side slip angle and a yaw angular velocity, takes driving signals provided by a driver and measured or estimated vehicle state information as input, and judges the stability of the vehicle. When the active safety system needs to intervene, a DDPG-AAD model trained offline by a strategy deep gradient algorithm is used to coordinately control AFS, ARS and DYC systems, precise and independent control of each wheel is realized, and thus the active safety driving stability of the vehicle is improved.

The invention belongs to the crossing field of an automobile active safety system and a high-precision gas sensing technology, and provides a vehicle-mounted alcohol system and method which can avoid interference of other ethanol-containing substances and only detect gas exhaled by a driver. The system defects of an existing vehicle-mounted alcohol detection system in the aspects of reliability, anti-interference performance and integration are overcome. On the basis of traditional alcohol detection and photoacoustic spectrum detection, the detection sensitivity, reliability and accuracy are improved, meanwhile, the device can be integrally applied to a vehicle-mounted alcohol detection system, and the requirement for rapidly and accurately detecting whether a driver drinks alcohol or not in a vehicle in real time is met.

The invention provides a tire-road adhesion coefficient real-time estimation method and a tire-road adhesion coefficient real-time estimation system fusing data and a model. The method comprises the following steps: acquiring historical time sequence data of vehicle motion, synchronously optimizing a main task and an auxiliary task and dynamically distributing loss weights by using an LSTM network with a multi-task training mechanism, thereby extracting road adhesion feature vectors in acceleration and wheel speed; and calculating the variance of the feature vector to evaluate the confidence coefficient, dynamically adjusting the observation noise covariance of the extended Kalman filter according to the confidence coefficient, and indirectly adapting to the Kalman gain weight. And inputting the feature as a virtual observed quantity into an extended Kalman filter, and performing state estimation in combination with a vehicle dynamics model. And finally, outputting a high-reliability adhesion coefficient real-time estimation value. According to the method, the feature learning ability of data driving and the state estimation robustness of model driving are effectively combined, accurate and adaptive road adhesion perception is realized, and the real-time state perception requirement of an advanced active safety system is met.

The invention relates to the technical field of traffic safety, and constructs a man-helmet-vehicle road cloud cooperative safety ecology. Aiming at the problems of low intelligent level, few safety configuration functions, difficulty in law enforcement management, inaccurate positioning of complex road conditions, afraid rescue of accidents and the like of a riding vehicle, the invention discloses a riding intelligent safety helmet and an active and passive safety guard system. The riding intelligent safety helmet comprises a helmet system which is linked with the vehicle for detection, control and traceability; an anti-collision AEB and an active safety system capable of preventing head upwarp, controlling the speed at a curve and slowly descending on a steep slope are arranged; the passive safety system has the functions of SOS alarm, complex site accurate positioning and passerby rescue responsibility-free evidence obtaining; the mobile and fixed law enforcement device is composed of a UHF RFID reader and a directional antenna array, and recognition is not affected by the environment. The riding end of the riding big data platform is mainly used for helmet and bicycle binding, transaction and safety education, and the cloud end is used for safe driving scoring, a theme data warehouse, a safety knowledge base and quality traceability. The system protects the riding safety from the source in a linkage cooperation and multi-dimensional data driving mode.

The invention relates to a full-automatic safety type natural gas kiln drying system. Comprising a PLC control unit, a combustion control module, a multi-type sensor assembly, a high-energy ignition device, a safety relief valve, a main control electromagnetic valve, an emergency cut-off valve, a pressure protection module, an alarm unit, a flow monitoring device, a manual adjusting valve, a manual throttling valve, a flexible connecting assembly, a gas injection pipe, an air supply unit and a rotary kiln body. Multi-sensor linkage control is integrated through a PLC intelligent center, a main and standby double-gas-supply redundant pipeline structure is innovatively adopted, a high-pressure / low-pressure double-protection mechanism and a flexible expansion joint design are combined, and full-closed-loop automatic control over the natural gas combustion process is achieved. An active safety system with flame detection interlocking ignition, pressure abnormity emergency cut-off and air volume dynamic matching is mainly constructed.

The invention discloses a vehicle active safety pre-detection method based on travel risk prediction, and relates to the technical field of vehicle active safety control, and the method comprises the steps: obtaining a historical travel data set of a vehicle, carrying out the division of road segments based on a road network topology, carrying out the risk feature extraction of the divided road segments, and constructing a road segment-risk feature mapping library; historical risk features of the road section are extracted from the mapping library, and the static features and the dynamic features of the road section are fused through a double-flow LSTM-attention mechanism model to obtain a road section risk probability; dividing the risk levels of the road sections according to the risk probability of the road sections, mapping the risk levels to an HSV color space, and transmitting the risk levels to a pre-check layer of the vehicle-mounted HUD; and calculating the triggering distance of the vehicle entering the predetermined road section, extracting risk level information from the pre-detection layer after the vehicle enters the triggering area, and triggering and loading the active safety pre-detection of the vehicle. According to the invention, the prediction capability and intervention capability of the vehicle on potential risks are improved, and the self-adaptability and intelligent level of a vehicle active safety system are enhanced.

The invention provides an active safety system based on an engineering machinery instrument panel. Comprising an instrument panel, a processor is arranged in the instrument panel, the processor is connected with cameras arranged around the whole machine equipment, and the cameras transmit shot pictures to the processor for synthesis; the processor is further connected with an engine controller and a whole vehicle controller, the engine controller is connected with a whole vehicle idling control module and a whole vehicle flameout control module, and the whole vehicle controller is connected with a whole vehicle hydraulic action control module and a whole vehicle walking control module. Through sound and color alarm modes of an engineering machinery instrument panel, fusion means such as a sensor technology, a graphic synthesis technology and a control technology are effectively utilized, the surrounding environment of operation of engineering machinery equipment is monitored in real time, accident risks are actively recognized, sudden accidents of an engineering equipment construction site are effectively reduced, and secondary damage is prevented.

This invention relates to the field of vehicle collision detection technology, and specifically discloses a vehicle collision detection method and system based on an improved dual-stream convolutional network, comprising the following steps: Step S1: Real-time acquisition of an original video stream containing the target vehicle via an onboard camera; Step S2: Preprocessing the original video stream to extract spatial frame sequences and optical flow frame sequences; Step S3: Inputting the spatial frame sequences into the spatial flow branch of the improved dual-stream convolutional network. This vehicle collision detection method and system based on the improved dual-stream convolutional network achieves a significant improvement in detection accuracy and extremely low end-to-end processing latency, providing sufficient warning and response time for active safety systems. Its dual-stream parallel processing and adaptive feature fusion design effectively balance computational efficiency and feature integrity, enabling the system to maintain a low false alarm rate in complex driving scenarios and clearly distinguish between emergency braking and real collision risks.

The present invention provides a vehicle stability integrated control method based on a dual stability envelope. This method constructs a dual stability envelope control framework using vehicle state soft constraints and actuator stability hard constraints. Vehicle instability conditions are mapped to actuator stability limits as actuator hard constraints. The dual stability envelope range is determined by combining vehicle state soft constraints with dynamic wheel posture updates and a stability boundary correction function, thereby balancing vehicle stability performance and controller computational complexity. The present invention proposes a wheel posture dynamic update method and a stability boundary correction function. The vehicle stability boundary is adjusted in real time based on driving conditions, driver input, and active safety system control input, improving the accuracy of the vehicle stability boundary description. A formula for actuator stability limits is derived based on the vehicle stability boundary. Vehicle instability conditions are mapped to actuator stability limits to determine the actuator stability envelope range. This prevents vehicle instability from a control input perspective and maximizes the vehicle stability boundary.

A vehicle window control method and apparatus, an electronic device, and a vehicle. The method comprises: in response to receiving a collision signal, starting a collision-induced window lowering determination execution step: acquiring vehicle speed information and the opening degree of each vehicle window; and in response to determining that there is a target window to be lowered and that a duration in which the vehicle speed information is less than or equal to a first preset vehicle speed reaches a preset duration, sending a window lowering request to a window driving apparatus corresponding to said target window, and controlling the window driving apparatus to adjust the opening degree of said target window to a first preset opening degree. The determination of conditions for triggering a collision-induced window lowering strategy is rational and effective, thereby enhancing the overall safety performance of vehicles, and further facilitating the optimization of an active safety system of modern vehicles.

Disclosed are systems, apparatus and methods for implementing an active safety system in an autonomous vehicle. An autonomous vehicle (100) may be travelling through an environment (190) external to the autonomous vehicle (100) along a trajectory (105). The environment (190) may include one or more objects that may potentially collide with the autonomous vehicle (100), such as static and / or dynamic objects, or objects that pose some other danger to passengers riding in the autonomous vehicle (100) and / or to the autonomous vehicle (100). An object (180) (e.g., an automobile) is depicted as having a trajectory (185), that if not altered (e.g., by changing trajectory, slowing down, etc.), may result in a potential collision (187) with the autonomous vehicle (100) (e.g., by rear-ending the autonomous vehicle (100)). The autonomous vehicle (100) may use a sensor system to sense (e.g., using passive and / or active sensors) the environment (190) to detect the object (180) and may take action to mitigate or prevent the potential collision of the object (180) with the autonomous vehicle (100).

The present invention discloses a threshold adjustment method and system for a vehicle active safety warning system. By collecting real-time vehicle driving data, driving environment data, driver status data, and active safety warning data, a vehicle driving safety warning dataset and a driver abnormal status warning dataset are constructed for warning mode categories coupled with different driving road scenarios and driving weather environments. Warning lag and warning advance threshold evaluation indicators are established for each warning category. The warning threshold is adjusted and judged based on the evaluation indicators, and the warning threshold is adjusted based on the evaluation results. The present invention can continuously quantitatively evaluate and adjust the warning threshold of the active safety system based on the warning data collected in real time during the warning system and vehicle driving, further improving the reliability and robustness of the active safety warning system.

The embodiment of the invention provides a control method and device for a vehicle headrest, an adjusting mechanism, a vehicle and a medium. The method comprises the steps that a starting signal of an active safety system of the vehicle is responded, and a trigger signal used for adjusting a headrest accompanying adjusting mechanism of the vehicle headrest is obtained; based on the trigger signal, controlling the headrest to move along with the adjusting mechanism and / or rotate the headrest to a preset position; the preset position is the position where the headrest is attached to the head of the passenger when the vehicle is collided and / or steers emergently. A starting signal of an active safety system serves as an input signal of a headrest accompanying adjusting mechanism, and the headrest accompanying adjusting mechanism with the moving and rotating functions is triggered in time to move and / or rotate a headrest to the position where the headrest is attached to the head of a passenger. Timely accompanying movement of the seat headrest and the head of a passenger in a potential rear-end collision accident is achieved, the phenomenon that the head of the passenger is out of position relative to the seat headrest is avoided, and the protection efficiency of the seat headrest on the head of the passenger is improved.

The present invention provides a bus braking energy recovery control method and system based on series control. The method monitors the brake pedal position and four-wheel speed through a linear Hall sensor and a magnetoelectric wheel speed sensor, and collects battery status data through a distributed battery status monitoring system. The Kalman filter is used to process the data to detect abnormal values, and a braking demand prediction model combining LSTM and GNN is constructed. The braking demand is predicted according to the real-time driving state. A non-dominated sorting genetic algorithm II is used for multi-objective dynamic optimization to determine the optimal braking torque distribution strategy. An adaptive slope calculation model is established according to the real-time load and road friction coefficient. The coasting brake intervention and exit rates are dynamically adjusted, a hierarchical energy recovery control strategy is established, and a priority mechanism is set to ensure that the optimal energy recovery strategy is selected when multiple conditions are met. In addition, a forced exit mechanism is set for situations such as activation of the active safety system and excessive battery temperature to ensure safety.

The utility model provides a square column-shaped roadblock target object for automobile active safety system testing, which relates to the technical field of automobile active safety system testing and comprises a target object body, the surface of the target object body is smooth and flat, the target object body has preset supporting force, four corners of the target object body are provided with chamfered edges, and the periphery of the target object body is provided with side walls along the axial direction. The working size of the side wall is matched with that of the chamfering edge; the outer surface of the target object body is coated with a coating; according to the utility model, the surface of the target object body is smooth and flat, and the surface of the target object body is coated with the coating, so that the target object body has good crashworthiness and wear resistance, and can keep stable performance in multiple tests; through the design shape of the square column of the target object body, an accurate parking test target can be provided for an automobile active safety system, and the consistency and reliability of test results are improved.

The utility model is suitable for the field of scooter controllers, and provides a scooter controller self-starting circuit, which comprises an MCU (Microprogrammed Control Unit), the MCU adopts a U5 chip, the model of the U5 chip is TMS320F28034PNT, and a pin 11, a pin 12, a pin 13, a pin 14, a pin 16, a pin 17 and a pin 18 of the U5 chip are respectively connected with an ADCINA7, an ADCINA6, an ADCINA5, an ADCINA4, an ADCINA3, a BrakeCheck, an ADCINA1 and an ADCINA0; according to the utility model, the safety and the reliability are enhanced: a set of active safety system is additionally arranged on the basis of the existing braking system of the scooter and the wheelchair, and when the electric door lock is not opened but the scooter slides, the controller can be automatically started to control a power supply, so that the controller works, the braking measure is triggered, the high-speed scooter sliding is effectively prevented, and the safety and the reliability of the whole scooter are greatly improved; the braking logic is perfected; and the power supply stability is improved.

The invention relates to the technical field of fuel gas. A gas active safety system with infrasonic wave turbulent flow comprises an electric control gas valve arranged on a gas pipeline and a microprocessor system. The electric control air valve is provided with a control signal input end, and a control output port of the microprocessor system is controllably connected with the control signal input end of the electric control air valve; the device further comprises a fuel gas concentration sensor system with infrasonic wave turbulent flow, the fuel gas concentration sensor system comprises an electromagnet, an electromagnetic movable part of the electromagnet is connected with a vibrating membrane, and an electromagnetic induction coil driving electromagnetic movable part with an infrasonic wave oscillating circuit driving electromagnet is connected with a vibrating membrane. And the vibration membrane promotes the airflow to flow through the fuel gas concentration sensor assembly, so that the sensitivity is improved, the reaction is quicker, and the fuel gas active safety system can make a quicker response under the condition that the concentration of the gas to be detected is lower.

This invention provides a method, system, device, and medium for uploading EDR data to the cloud for intelligent driving vehicles, belonging to the field of intelligent driving vehicle technology. The method includes: real-time acquisition of EDR data and writing it into a buffer; continuous monitoring of the vehicle's dynamic state and the working state of the active safety system; sending a trigger command when a collision or potential collision event meeting preset conditions is detected; according to the trigger command, storing the EDR data in the buffer for a first preset time period before the trigger command occurs on a disk, and continuously storing the EDR data for a second preset time period after the trigger command occurs on a disk, performing data compression during the disk storage process; and uploading the stored EDR data to the cloud. This invention achieves autonomous and controllable uploading of EDR data to the cloud, enabling efficient data retrieval, and allowing for in-depth analysis and playback of accidents based on cloud platform data management combined with real-time data acquired at the time of EDR triggering, without relying on locally stored data.

The invention relates to the technical field of intelligent traffic systems and active safety, and provides an active safety system for the field of automatic driving, including but not limited to an active safety system for improving safety or reducing safety accident loss to the greatest extent. Specifically, the system provides system-based security metric functionality by integrating one or more elements of a vehicle component, a road component, a cloud component, a pedestrian component, and a map component. The security metric functions include one or more of collision detection, collision prediction, collision early warning, collision avoidance, collision impact mitigation, and emergency response functions. According to the invention, all-around and multi-level active safety guarantee can be provided for safe operation of the automatic driving vehicle in a complex dynamic traffic environment.

The invention relates to the technical field of intelligent traffic management and planning, provides a vehicle-centered enterprise-level active safety system, and aims to improve safety or minimize loss caused by safety events. More specifically, the vehicle-centric enterprise-level active safety system provides system-based safety measures by integrating one or more vehicle components, pedestrian components, and map components. The security measures include one or more functions, such as management and service, collision detection, collision prediction, collision early warning, collision avoidance, collision impact reduction, and emergency response functions.

The invention discloses a multi-parameter sensing intelligent tire based on a three-way acceleration sensor. The multi-parameter sensing intelligent tire comprises a tire body, a low-power-consumption three-way acceleration sensor module and a signal processing and wireless communication module. The acceleration sensor can collect acceleration signals of a tire in the driving process in real time, and multi-dimensional recognition is carried out on parameters such as unevenness, adhesion coefficient and lateral force of a road surface in combination with an improved mixed Lagrange-Euler algorithm and a deep learning method. Collected multi-parameter road condition information is sent to a vehicle terminal in a wireless communication mode, and intelligent sensing of road surface types and states is achieved. The problems that in the prior art, the tire and road surface contact state information is difficult to obtain, and response is slow are solved, and the safety and stability of a vehicle auxiliary driving or unmanned driving system under the complex road condition are remarkably improved. The device has the advantages of high integration level, low energy consumption, high recognition accuracy and the like, and is suitable for automatic driving and active safety systems of various intelligent vehicles.

The invention relates to the technical field of intelligent driving and active safety, and provides an enterprise-oriented vehicle cloud active safety system, which is used for improving the vehicle operation safety and minimizing the safety event loss. The system comprises a hardware assembly composed of a vehicle assembly and a cloud assembly, and a software assembly composed of a management and service subsystem, a collision detection subsystem, a collision prediction subsystem, a collision early warning subsystem, a collision avoidance subsystem, a collision influence reduction subsystem and an emergency response subsystem. A pedestrian component and a map component can also be included. The enterprise-oriented vehicle cloud active safety system is used for realizing one or more of the following functions: collision detection, collision prediction, collision early warning, collision avoidance, collision influence reduction and emergency response. According to the invention, comprehensive active safety protection can be provided for an enterprise-level intelligent driving system, the adaptive capacity and accident coping capacity of the CAV / AV in a complex traffic environment are enhanced, and the overall level of road traffic safety is improved.