Intelligent connected vehicle based complex urban environment perception and collision warning system

Abstract

The application discloses a complex urban environment sensing and collision warning system based on intelligent networked vehicles and relates to the technical field of electric vehicle control. The complex urban environment sensing and collision warning system based on intelligent networked vehicles comprises an obstacle positioning and identification analysis module, a data fusion and synchronous analysis module and a collision risk warning analysis module. In the obstacle sensing stage of the intelligent networked platform, the positioning and identification analysis of the data sensing process is carried out to determine whether radar wave adjustment optimization is to be performed, then in the data processing stage, data fusion and synchronous analysis is carried out to determine whether data capacity window optimization is to be performed, and finally in the collision risk warning stage, collision risk assessment analysis is carried out to determine whether filter cache optimization is to be performed, so that the problem of the increase of the collision warning false alarm rate caused by the obstacle sensing delay of the intelligent networked vehicle in the peak period road section in the prior art is solved.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle control technology, and in particular to a complex urban environment perception and collision warning system based on intelligent connected vehicles. Background Technology

[0002] In the field of Intelligent Connected Vehicles (ICVs), complex urban environment perception and collision warning systems are among the core technologies for ensuring driving safety and enhancing the driving experience. Urban environments, characterized by rapid changes, diverse obstacles, and complex traffic rules, place extremely high demands on vehicles' perception and decision-making capabilities. Intelligent connected vehicles integrate advanced sensors, big data processing, and artificial intelligence algorithms to achieve accurate perception and risk assessment of the surrounding environment. This allows them to issue warnings to the driver before a potential collision occurs and even automatically take evasive action in emergencies, such as emergency braking or steering, thereby significantly reducing the traffic accident rate.

[0003] Specifically, static or dynamic information in the urban environment is collected through multi-source heterogeneous sensors. For example, vehicle-mounted cameras are used to collect visible light images of the road segment to detect obstacles and warning signs such as vehicles, pedestrians, and traffic signs, and the road area is divided into regions. Millimeter-wave radar is used to detect obstacles, such as distance, speed, and angle measurement, and the complex urban environment around the vehicle is transformed into a 3D point cloud form in the intelligent connected platform. When the vehicle is close, ultrasonic radar is used to detect nearby obstacles in more detail, such as the distance between vehicles in front and behind during traffic jams and sudden pedestrians crossing the road. In terms of data fusion, while ensuring the integrity of multi-source data streams, the acquired urban environmental data is matched and fused, and the recognition probability is evaluated with confidence. In terms of risk assessment, a reinforcement learning model is used to calculate the collision risk based on the acquired environmental data, and the corresponding pre-defined warning strategy is output in the model. The warning method (such as sound prompts, visual prompts, touch prompts, etc.) is selected according to the collision risk calculation results and risk level. If necessary, the platform can intervene forcibly, such as forced braking.

[0004] For example, Chinese invention patent CN118182510B discloses a driving safety early warning system applicable to intelligent connected vehicles, including: a driver assessment module that monitors the behavior and physical condition of the driver in the vehicle during the detection period; a road environment perception module that analyzes the road risk conditions of the perceived road section; and a system that captures and analyzes abnormal vehicle operation during driving, and sends the corresponding alarm information obtained and generated driver warning signals, road environment perception warning signals, or vehicle operation abnormality warning signals to the driving warning terminal.

[0005] For example, Chinese invention patent with announcement number CN118387135B discloses an intelligent connected vehicle autonomous driving detection and control method and its control system, including: continuously reinforcing learning to plan the target trajectory based on steering decision information and lane information, and judging whether to adjust the target trajectory according to the obtained front wheel steering angle loss value, and adjusting the planned target trajectory in a timely manner in combination with the front wheel steering angle loss during trajectory planning.

[0006] The above-mentioned technology has at least the following technical problems: In existing technologies, during peak hours, traffic volume increases exponentially, roads are densely packed with vehicles, and pedestrians and non-motorized vehicles frequently weave through the traffic with complex and rapidly changing trajectories. Intelligent connected vehicles rely on multiple sensors to perceive their surroundings. Faced with such a complex scenario, these sensors need to collect massive amounts of data, undoubtedly placing enormous pressure on the data collection process and increasing its time consumption. After data collection, the data must be transmitted to the onboard computing platform via the network. During peak hours, the simultaneous transmission of data by numerous vehicles can easily cause network congestion, leading to queuing, packet loss, and other issues during transmission, further exacerbating latency. Upon receiving the data, the onboard computing platform must fuse, analyze, and identify the multi-source heterogeneous data to obtain accurate environmental information and conduct collision risk assessments. The massive amount of data places a heavy processing burden on the computing platform, increasing processing time. These delays cause discrepancies between the environmental information acquired by the system and the actual scenario, which can easily lead to misjudgments during collision risk assessment. Safe scenarios may be mistakenly identified as dangerous scenarios, thus triggering collision warnings and increasing the false alarm rate. This presents a problem where the false alarm rate of collision warnings is increased due to the delay in obstacle perception in intelligent connected vehicles on roads during peak hours. Summary of the Invention

[0007] To address the technical problem of increased false alarm rates in collision warnings caused by delayed obstacle perception in intelligent connected vehicles during peak hours, this invention provides a complex urban environment perception and collision warning system based on intelligent connected vehicles. The technical solution is as follows: A complex urban environment perception and collision warning system based on intelligent connected vehicles is provided. This system includes: an obstacle localization and identification analysis module, used to perform localization and identification analysis on the obstacle perception process of the intelligent connected platform based on acquired obstacle perception data, to determine whether radar wave adjustment optimization is needed. Localization and identification analysis quantifies the accuracy of the intelligent connected platform's perception of obstacle positions in collision risk areas, while radar wave adjustment optimization improves the platform's obstacle perception capability; and a data fusion and synchronization analysis module, used to perform data fusion and synchronization analysis on the obstacle data processing process of the intelligent connected platform based on acquired obstacle synchronization identification time. The system determines whether to perform data capacity window optimization. Data fusion synchronization analysis is used to quantify the time alignment of obstacle data processing by multi-source sensors in the intelligent connected platform. Data capacity window optimization is used to improve the intelligent connected platform's ability to fuse and synchronize obstacle data. The collision risk warning module is used to perform collision risk assessment and analysis on the obstacle movement trajectory of the intelligent connected platform based on the acquired collision risk prediction data to determine whether to perform filtering and caching optimization. Collision risk assessment and analysis is used to quantify the accuracy of the intelligent connected platform's prediction of the collision risk between the intelligent connected vehicle and the obstacle. Filtering and caching optimization is used to improve the accuracy of collision risk prediction and the reliability of collision risk warning.

[0008] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: 1. Through the obstacle perception process of the intelligent connected platform, perform positioning and identification analysis, and determine whether to perform radar wave adjustment and optimization based on the analysis results to improve the obstacle edge feature extraction capability and improve the positioning resolution of low-speed and small obstacles; in the obstacle data processing stage, perform data fusion and synchronous analysis of the processing process, and determine whether to optimize the data capacity window based on the analysis results to ensure the continuity of obstacle motion trajectory prediction; in the collision risk warning stage, perform collision risk assessment and analysis of obstacle motion trajectory, and determine whether to perform filtering and caching optimization based on the analysis results to smooth out trajectory prediction abrupt changes caused by noise interference.

[0009] 2. By fusing trajectory prediction delay and ranging delay into a collision risk prediction index through harmonic averaging, this method avoids the feature suppression effect caused by neglecting dimensional mismatch in traditional methods and overcomes the problem of hysteresis error accumulation that may be caused by linear superposition. This method uses a dynamic balancing mechanism to adaptively adjust the weights of the two types of delays on the system according to the complexity of the traffic scenario: strengthening the sensitivity constraint of ranging delay in dense traffic scenarios, and focusing on the fault tolerance correction of prediction delay under trajectory change. Compared with the traditional fixed weight fusion method, the harmonic averaging mechanism improves the system's feedforward compensation capability for the two types of delays, so that the collision risk prediction index can ensure real-time performance while also having the ability to dynamically align deviations.

[0010] 3. By integrating the antenna array response time with the multipath interference ranging time and radar reception response time, this method, compared with the limitations of existing technologies in managing multipath scattering noise and response inconsistency interference, weakens the time delay mismatch problem caused by traditional sensing by influencing the influence of environmental correction values ​​and inherent characteristics on feature values. Compared with the conventional linear superposition compensation strategy, the three-level difference degree coupling mechanism can not only suppress the path tracking dispersion effect caused by strong reflectors, but also enhance the adaptive compensation capability in dynamic scenes through hierarchical weight allocation. In typical complex scenarios such as dense pedestrian traffic, it avoids the risk of moving target trajectory adhesion and feature point matching inaccuracy that may be caused by traditional dimensional normalization methods. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A schematic diagram of the structure of a complex urban environment perception and collision warning system based on intelligent connected vehicles provided in an embodiment of the present invention; Figure 2 The execution flowchart of the obstacle localization and identification analysis module provided in the embodiment of the present invention; Figure 3 The execution flowchart of the data fusion synchronization analysis module provided in this embodiment of the invention; Figure 4 The execution flowchart of the collision risk warning and analysis module provided in the embodiment of the present invention is shown. Detailed Implementation

[0013] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0014] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0015] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0016] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0017] This invention provides a complex urban environment perception and collision warning system based on intelligent connected vehicles, such as... Figure 1 The diagram shows a structural schematic of a complex urban environment perception and collision warning system based on intelligent connected vehicles. This system includes the following modules: an obstacle localization and identification analysis module, used during the obstacle perception phase of the intelligent connected platform, performs localization and identification analysis on the obstacle perception process based on the acquired obstacle perception data to determine whether radar wave adjustment optimization is needed. The localization and identification analysis quantifies the accuracy of the intelligent connected platform's perception of obstacle positions in collision risk areas, while radar wave adjustment optimization improves the platform's obstacle perception capability. A data fusion and synchronization analysis module, used during the obstacle data processing phase of the intelligent connected platform, analyzes the obstacle data processing based on the acquired obstacle synchronization recognition time. The process involves data fusion and synchronization analysis to determine whether data capacity window optimization is necessary. This analysis quantifies the time alignment of obstacle data processing by multi-source sensors in the intelligent connected platform, while data capacity window optimization improves the platform's ability to fuse and synchronize obstacle data. The collision risk warning module, during the collision risk warning phase of the intelligent connected platform, performs collision risk assessment and analysis on the obstacle trajectory based on acquired collision risk prediction data to determine whether filtering and caching optimization is needed. This assessment quantifies the accuracy of the intelligent connected platform's prediction of collision risks between the vehicle and obstacles, while filtering and caching optimization improves the accuracy of collision risk prediction and the reliability of collision risk warnings.

[0018] In this embodiment, the obstacle perception stage focuses on improving perception capabilities. Through obstacle localization and identification analysis, the platform quantifies the accuracy of its positional perception of obstacles in collision risk areas, accurately determining whether radar wave adjustment and optimization are necessary. This stage effectively solves the positioning deviation problem caused by obstacle occlusion and signal interference in urban environments. After optimization, the radar wave coverage and resolution are significantly enhanced, reducing the vehicle's spatial positional perception error of surrounding obstacles and providing reliable basic data for subsequent trajectory prediction.

[0019] The data processing stage enhances data fusion efficiency by simultaneously analyzing and quantifying the time alignment of multiple sensor sources (such as cameras and millimeter-wave radar) and dynamically adjusting the data capacity window. This stage overcomes the bottleneck of asynchronous data from heterogeneous sensors, ensuring the real-time nature of dynamic parameters such as obstacle speed and direction, and avoiding false alarms caused by data misalignment.

[0020] The collision warning phase enhances the reliability of risk assessment, while collision risk evaluation and analysis quantify the accuracy of collision prediction. Filtering and caching optimization improves trajectory prediction stability by suppressing noise signals. This phase shortens the system's response time to sudden collision scenarios and reduces the false alarm rate.

[0021] This example systematically illustrates the system's advantages from three aspects: functional implementation, technological breakthroughs, and application effects. In the obstacle perception stage, the accuracy of position perception is improved through positioning analysis and radar optimization. In the data processing stage, synchronous analysis and window adjustment break through the timing bottleneck and ensure the real-time performance of dynamic parameters. In the collision warning stage, evaluation analysis and filtering optimization shorten the response time and reduce the false alarm rate, ultimately achieving efficient and reliable collaboration across the entire "perception-processing-early warning" chain.

[0022] Furthermore, based on the acquired obstacle perception data, the obstacle perception process of the intelligent connected platform is located, identified, and analyzed. The specific steps are as follows: at the end of the obstacle perception period, the acquired obstacle perception data is compared with the preset obstacle perception data in the database. At the same time, the obstacle perception data correction value is combined to correct the comparison results of each difference, obtain the obstacle perception data score, and perform coupling processing to obtain the obstacle perception impact index.

[0023] The obstacle perception data includes antenna array response time, multipath interference ranging time, and radar reception response time. The obstacle perception data correction values ​​include antenna array response time correction values, multipath interference ranging time correction values, and radar reception response time correction values. The obstacle perception data scores include antenna array response time scores, multipath interference ranging time scores, and radar reception response time scores. The preset obstacle perception data includes preset antenna array response time, multipath interference ranging time, and radar reception response time. The obstacle perception impact index is used to quantify the degree of impact of obstacle perception data on the perception capabilities of the intelligent connected platform.

[0024] Specifically, antenna array response time reflects the time required for the antenna array to respond from receiving a signal during obstacle perception. The shorter the response time, the faster the antenna array can react to obstacle signals, which helps to obtain information such as the position and speed of obstacles in a timely manner. Multipath interference ranging time reflects the time required for the intelligent connected platform to complete accurate ranging operations in an environment of multipath interference (where signals encounter obstacles during propagation and are reflected or refracted, resulting in the receiver receiving signals from multiple different paths). Multipath interference increases the complexity and uncertainty of ranging, and this time reflects the ability of the intelligent connected platform to overcome multipath interference and perform accurate ranging. Radar reception response time reflects the time required for millimeter-wave radar to complete signal processing and provide an effective response after receiving a signal reflected from an obstacle. Millimeter-wave radar has high resolution and good anti-interference capabilities, and its reception response time directly affects the speed and accuracy of the intelligent connected platform in perceiving the detailed features of obstacles.

[0025] The specific expression for the antenna array response time fraction W1 is: In the formula, W1 represents the antenna array response time fraction corresponding to the end of the obstacle perception period of the intelligent connected platform, C1 represents the antenna array response time correction value, and W AR W represents the antenna array response time at the end of the obstacle perception period for the intelligent connected platform. AR1 This indicates the preset antenna array response time.

[0026] The specific expression for the multipath interference ranging duration fraction W2 is: In the formula, W2 represents the multipath interference ranging duration fraction corresponding to the end of the obstacle perception period for the intelligent connected platform, C2 represents the multipath interference ranging duration correction value, and W... MI W represents the multipath interference ranging duration corresponding to the end of the obstacle perception period for the intelligent connected platform. MI1 This indicates the preset multipath interference ranging time.

[0027] The specific expression for the radar receive response time fraction W3 is: In the formula, W3 represents the radar reception response time fraction corresponding to the end of the obstacle perception period of the intelligent connected platform, C3 represents the radar reception response time correction value, and W MR W represents the radar reception response time at the end of the obstacle perception period for the intelligent connected platform. MR1 This indicates the preset radar reception response time.

[0028] Obstacle perception impact index P D The specific expression is: In the formula, P DThis indicates the obstacle perception impact index corresponding to the end of the obstacle perception period for the intelligent connected platform.

[0029] In this embodiment, the antenna array response time, multipath interference ranging time, and radar reception response time are all acquired through a timer built into the intelligent connected platform. The preset antenna array response time, multipath interference ranging time, and radar reception response time are represented by the average of the historical antenna array response time, historical multipath interference ranging time, and historical radar reception response time acquired by the intelligent connected platform at the end of the historical obstacle perception period. The correction values ​​for antenna array response time, multipath interference ranging time, and radar reception response time are preset in the database to indicate the degree of influence of the antenna array response time, multipath interference ranging time, and radar reception response time on the acquisition of obstacle perception impact indicators. In the formula, the database stores the corresponding correction values ​​for each parameter, and the values ​​of the three correction values ​​are all in the range of 0-1, and the sum of them is equal to 1. In practical applications, the corresponding correction values ​​can be obtained by inputting the real-time antenna array response time, multipath interference ranging time, and radar reception response time, providing accurate quantitative data for evaluating obstacle perception impact indicators.

[0030] It is important to note that the impact of obstacle perception on the overall signal processing time increases with the increase of antenna array response time, multipath interference ranging time, and radar reception response time. Specifically, the antenna array response time increases with the increase of multipath interference ranging time because it requires more time for dynamic beamforming optimization to suppress stray reflections. The radar reception response time is simultaneously constrained by the former two factors. When the antenna array beam decreases, the response time increases because the computational load caused by multipath interference increases, leading to an increase in multipath interference ranging time. At the same time, the radar needs to maintain the signal-to-noise ratio of the echo signal, thus extending the overall signal reception processing time.

[0031] Considering the aforementioned mutual influence mechanisms, a more comprehensive understanding of the relationship between obstacle perception impact indicators and various variables can be achieved. Accurate evaluation of these relationships is crucial in the process of acquiring obstacle perception impact data. By optimizing antenna array response time, multipath interference ranging time, and radar reception response time, the impact of obstacle perception can be effectively reduced, improving the accuracy and stability of obstacle perception. This makes the acquired obstacle perception data more closely reflect actual effects, thereby effectively solving the problem of unreliable obstacle perception capabilities of intelligent connected platforms.

[0032] like Figure 2The diagram shows the execution flowchart of the obstacle location identification and analysis module provided in this embodiment of the invention. If the acquired obstacle perception impact index is not greater than the preset value in the database, data fusion and synchronous analysis are performed. If it is greater, the radar wave transmission power is adjusted and the obstacle perception impact index is output. It is then determined whether the newly acquired index is not greater than the preset value in the database. If not, data fusion and synchronous analysis are performed. If so, the beamwidth is adjusted and the newly acquired index is compared with the preset index in the database. If it is greater than the preset value in the database, an obstacle perception risk warning is issued. If it is not greater, data fusion and synchronous analysis are performed.

[0033] Further understanding is needed regarding the specific steps for determining whether radar wave adjustment and optimization should be performed: The acquired obstacle perception impact index is compared with the preset obstacle perception impact index in the database. If the acquired obstacle perception impact index is greater than the preset obstacle perception impact index in the database, it is recorded as obstacle perception unqualified and radar wave adjustment and optimization is performed. Radar wave adjustment and optimization means improving the intelligent connected platform's obstacle perception response by adjusting the radar wave transmission power and beamwidth. If the acquired obstacle perception impact index is not greater than the preset obstacle perception impact index in the database, it is recorded as obstacle perception qualified and data fusion and synchronous analysis is performed.

[0034] The radar wave adjustment and optimization process involves the following steps: The radar wave transmission power adjustment value, mapped from the acquired obstacle perception impact index deviation in the database, is used to quantify the specific degree to which the radar wave intensity needs to be increased, thereby reducing the probability of obstacle perception errors due to signal interference. The obstacle perception impact index deviation represents the difference between the acquired obstacle perception impact index and the preset obstacle perception impact index in the database. The preset obstacle perception impact index is represented by the summation and averaging of historical obstacle perception impact indices from the intelligent connected platform at the end of the obstacle perception period. After the radar wave transmission power adjustment is completed, the intelligent connected platform sends a transmission power verification command to the preset personnel, prompting them to conduct a positioning and identification verification test based on the adjusted radar wave transmission power. After the positioning and identification verification test is completed, the obstacle perception process is repeated. If the newly acquired obstacle perception impact index is greater than the preset obstacle perception impact index, an obstacle perception warning is issued; otherwise, the radar wave transmission power adjustment is completed, and data fusion and synchronous analysis are performed.

[0035] It is important to understand that radar wave adjustment optimization also includes beamwidth optimization. The specific steps are as follows: based on the deviation of the obstacle perception impact index that is re-acquired at the end of the re-performed obstacle perception process, a beamwidth adjustment value is mapped in the database to quantify the degree to which the beamwidth needs to be reduced to reduce signal interference caused by dense obstacles; after the beamwidth adjustment, if the re-acquired obstacle perception impact index is not greater than the preset obstacle perception impact index in the database, the radar wave adjustment optimization is completed and data fusion synchronous analysis is performed; otherwise, an obstacle perception risk warning is issued.

[0036] In this embodiment, an adaptive power control algorithm is used to increase the power intensity and beamwidth in a timely manner based on the acquired current transmit power and beamwidth, thereby improving the intelligent connected platform's ability to perceive obstacles until the obstacle perception is qualified. At the same time, historical power intensity and beamwidth are used as sample data and input into a reinforcement learning model. Based on the adaptive power control algorithm, a transmit power-beamwidth reinforcement learning model is obtained through training. The acquired power intensity and beamwidth are input into the transmit power-beamwidth reinforcement learning model, and the corresponding adjustment values ​​of transmit power and beamwidth are output.

[0037] This example enhances signal penetration and anti-interference capabilities, compensates for the attenuation of the perceived signal caused by abrupt changes in obstacle material or environmental noise, and reduces the probability of false detection and missed detection. Beamwidth optimization suppresses the delay caused by noise signal interference, improves the target resolution accuracy and boundary recognition clarity in dense obstacle groups, ensures precise matching of power and beam parameter adjustment, and avoids signal saturation or blind zone coverage caused by over-optimization.

[0038] like Figure 3 The diagram shows the execution flowchart of the data fusion synchronization analysis module provided in this embodiment of the invention. It determines whether the acquired obstacle synchronization recognition time is greater than a preset value. If not, a collision risk assessment analysis is performed. If so, the obstacle synchronization recognition time is obtained by adjusting the data compression ratio, and compared with the preset value in the database. If not satisfied, the buffer capacity is adjusted and the obstacle synchronization recognition time is reacquired. It then determines whether the acquired value meets the preset requirements. If not, the parallel processing window size is adjusted, and the obstacle synchronization recognition time obtained on this basis is determined to be greater than the preset value in the database. If so, a data fusion warning is issued; if satisfied, a collision risk assessment analysis is performed.

[0039] Further understanding is needed regarding the specific steps for determining whether data capacity window optimization should be performed: The acquired obstacle synchronization recognition time is compared with the preset obstacle synchronization recognition time in the database. The obstacle synchronization recognition time represents the time from when the intelligent connected platform receives environmental perception data to when it completes obstacle localization through fusion, and is obtained through a built-in timer. If the acquired obstacle synchronization recognition time is greater than the preset obstacle synchronization recognition time in the database, it is recorded as unqualified synchronization recognition and data compression optimization is performed. Data compression optimization means adjusting the data compression ratio to improve the intelligent connected platform's response to obstacle location recognition. If the acquired obstacle synchronization recognition time is not greater than the preset obstacle synchronization recognition time in the database, it is recorded as qualified synchronization recognition and collision risk assessment analysis is performed.

[0040] The data compression optimization process is as follows: Based on the acquired obstacle synchronization recognition time deviation, a corresponding data compression ratio adjustment value is mapped in the database to reduce the recognition deviation caused by the ambiguity of environmental perception data. The obstacle synchronization recognition time deviation represents the difference between the acquired obstacle synchronization recognition time and the preset obstacle synchronization recognition time in the database. The preset obstacle synchronization recognition time is represented by the sum and average of the historical obstacle synchronization recognition times at the end of the obstacle data processing period of the historical intelligent connected platform. After the data compression ratio is adjusted, if the newly acquired obstacle synchronization recognition time is greater than the preset obstacle synchronization recognition time in the database, capacity window optimization is performed; otherwise, data capacity window optimization is completed and collision risk assessment analysis is performed.

[0041] The capacity window optimization is as follows: Based on the obstacle synchronization recognition time deviation re-acquired after data compression optimization, a buffer capacity adjustment value is mapped in the database. This value is used to quantify the degree to which the buffer capacity needs to be increased, thereby improving the integrity of the intelligent connected platform's recognition data and thus improving recognition accuracy. After the buffer capacity adjustment is completed, if the re-acquired obstacle synchronization recognition time is not greater than the preset obstacle synchronization recognition time in the database, the data capacity window optimization is completed. Otherwise, based on the obstacle synchronization recognition time deviation, a corresponding parallel processing window adjustment value is mapped in the database. This value is used to quantify the degree to which the size of the parallel processing window needs to be increased, reducing task queuing delay while increasing the processing speed of multi-parallel threads. After the parallel processing window adjustment, if the re-acquired obstacle synchronization recognition time is greater than the preset obstacle synchronization recognition time in the database, a data fusion warning is issued. Otherwise, the data capacity window optimization is completed and a collision risk assessment analysis is performed.

[0042] In this embodiment, the data compression ratio is adaptively adjusted based on the current data compression ratio obtained by the statistical optimization algorithm until the synchronous identification is qualified. At the same time, the historical data compression ratio is used as sample data and input into the convolutional neural network to train the data compression-neural network model based on the statistical optimization algorithm. The obtained data compression ratio is input into the data compression-neural network model to output the corresponding data compression ratio adjustment value.

[0043] Similarly, the time series predictive control algorithm increases the capacity and window size in a timely manner based on the current buffer capacity and parallel processing window size until the synchronization recognition is qualified. The historical buffer capacity and parallel processing window size are input into the convolutional neural network and trained based on the time series predictive control algorithm to obtain the buffer capacity-parallel processing neural network model. The obtained buffer capacity and parallel processing window size are input into the model and the corresponding adjustment value is output.

[0044] This example reduces the processing resources consumed by non-critical data (such as the sky) by compressing data transmission and parsing latency. Secondly, the buffer expansion mechanism balances data integrity constraints and real-time requirements, and ensures the alignment of multiple data streams in terms of latency through elastic memory allocation, improving the coherence and smoothness of obstacle space extraction. The dynamic expansion of the parallel processing window optimizes the scheduling of multimodal data processing tasks, reduces queuing latency, and improves the real-time response speed in high-density obstacle scenarios. Adaptively matching the data fusion load fluctuations in complex urban environments ensures that the synchronous recognition time is consistently lower than the preset value, providing high-confidence time-synchronization feature input for collision risk assessment.

[0045] Furthermore, based on the acquired collision risk prediction data, a collision risk assessment and analysis is performed on the obstacle movement trajectory of the intelligent connected platform. Specifically, the acquired trajectory prediction delay time is compared with the preset trajectory prediction delay time in the database to obtain a trajectory prediction delay time score. At the same time, the acquired ranging delay time is compared with the preset ranging delay time in the database to obtain a ranging delay time score. The obtained ranging delay time score and trajectory prediction delay time score are harmonicly averaged to obtain a collision risk prediction index. The collision risk prediction index is used to quantify the impact of collision risk prediction data on the risk warning capability of the intelligent connected platform.

[0046] The collision risk prediction data includes trajectory prediction delay and distance measurement delay, both acquired through a timer built into the intelligent connected platform. The trajectory prediction delay represents the delay in predicting the future trajectory based on the target's historical state (such as position, speed, acceleration, and direction of motion). The shorter this delay, the lower the delay in the intelligent connected platform's ability to predict the activity state of obstacles in subsequent periods, which is beneficial for the platform to react quickly. The distance measurement delay represents the delay in measuring the distance between the obstacle and the vehicle. The shorter this delay, the lower the delay in the intelligent connected platform's measurement of the distance between the obstacle and the vehicle, which is beneficial for the platform's risk assessment.

[0047] In this embodiment, the preset trajectory prediction delay and ranging delay are represented by the sum of the historical trajectory prediction delay and ranging delay at the end of the collision risk warning period of the historical intelligent connected platform. In this example, the increase in trajectory prediction delay will cause the cumulative error of motion state to rise, requiring a longer ranging time to complete trajectory verification, resulting in an extension of ranging delay. At the same time, the aggravation of ranging delay reduces the update frequency of original distance information, and the trajectory prediction model needs to rely on more historical data interpolation to compensate for the lack of real-time performance, further amplifying the prediction delay.

[0048] The assessment of trajectory prediction delay score enhances the sensitivity to sudden lane changes or rapidly decelerating obstacles by shortening the time loss between historical feature extraction and future trajectory generation; the assessment of ranging delay score reduces data processing redundancy in distance measurement, enhancing the timeliness and continuity of dynamic obstacle ranging data; the coupled collision risk prediction index reflects the comprehensive risk situation; it can quickly capture urgent collision threats and accurately identify the evolution trend of potential risks.

[0049] like Figure 4 The diagram shows the execution flowchart of the collision risk warning analysis module provided in this embodiment of the invention. It determines whether the obtained collision risk prediction index meets the requirements, i.e., the obtained value is greater than the preset value. If it meets the requirements, a collision risk warning is issued. If it does not meet the requirements, the mean filtering intensity is adjusted and it is determined whether the collision risk prediction index obtained on this basis meets the conditions. If it does not meet the requirements, the historical data cache length is adjusted and it is determined whether the new index is greater than the preset value in the database. If it is greater than the preset value, a collision risk warning is issued. If it is not greater than the preset value, a "No collision risk instruction" is issued.

[0050] Further understanding is needed regarding the steps for determining whether to perform filter cache optimization: If the acquired collision risk prediction index is greater than the preset collision risk prediction index in the database, a collision risk is identified and a collision risk warning is issued; otherwise, mean filter strength optimization is performed. Specifically, the mean filter strength adjustment value is obtained by mapping the deviation of the acquired collision risk prediction index in the database. This value is used to quantify the degree to which the Kalman filter strength needs to be increased to reduce the deviation of the target prediction due to noise interference. After the filter strength adjustment, if the re-acquired collision risk prediction index is greater than the preset collision risk prediction index in the database, a collision risk warning is issued; otherwise, the cache length is adjusted. The collision risk prediction index deviation represents the difference between the acquired collision risk prediction index and the preset collision risk prediction index in the database. The preset collision risk prediction index is represented by the sum and average of the historical collision risk prediction indices at the end of the historical intelligent connected vehicle platform's collision risk warning period.

[0051] The specific process of cache length adjustment is as follows: Based on the collision risk prediction index deviation re-acquired after the filter strength adjustment, the historical data cache length adjustment value is mapped in the database. This value is used to quantify the current need to increase the historical cache length in order to reduce the deviation and misjudgment caused by short-term data fluctuations. After the cache length is adjusted, if the re-acquired collision risk prediction index is greater than the preset collision risk prediction index in the database, a collision risk warning is issued; otherwise, a command indicating that there is currently no collision risk is sent.

[0052] In this embodiment, the trajectory prediction algorithm increases the filtering strength and the reading buffer length in a timely manner based on the current obstacle motion state and buffer reading length until the collision risk prediction is completed. At the same time, the historical buffer length and filtering strength are used as sample data and input into the recurrent neural network. The neural network model of filtering strength-buffer length is trained based on the trajectory prediction algorithm. The obtained buffer length and filtering strength are input into the neural network model of filtering strength-buffer length and finally output the corresponding buffer length and filtering strength adjustment values.

[0053] It's important to understand that sensors in complex urban environments (such as during peak hours) are susceptible to noise, obstructions, or short-term data fluctuations, leading to falsely high (false alarms) or falsely low (missed alarms) instantaneous values. The core purpose of this optimization step is to address the issues of "false instantaneous value judgment" and "data fluctuation interference" in conventional collision risk assessment. Through a two-layer optimization loop of "filtering-caching," both real-time performance and prediction stability are ensured, ultimately achieving a leap from instantaneous value judgment to spatiotemporal continuity analysis in collision risk assessment, thereby effectively reducing the false alarm rate of collision warnings for intelligent connected vehicles during peak hours.

[0054] It should be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0055] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0056] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0057] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0058] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0059] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0060] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0061] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0062] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0063] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0064] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0065] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A complex urban environment perception and collision warning system based on intelligent connected vehicles, characterized in that, include: Obstacle localization and identification analysis module, data fusion and synchronization analysis module, and collision risk early warning analysis module; The obstacle location identification and analysis module is used to perform location identification and analysis on the obstacle perception process of the intelligent connected platform based on the acquired obstacle perception data, so as to determine whether radar wave adjustment and optimization should be performed. The location identification and analysis is used to quantify the accuracy of the intelligent connected platform in the intelligent connected vehicle in perceiving the position of obstacles in the collision risk area. The radar wave adjustment and optimization is used to improve the obstacle perception capability of the intelligent connected platform. The data fusion synchronization analysis module is used to perform data fusion synchronization analysis on the obstacle data processing process of the intelligent connected platform based on the acquired obstacle synchronization recognition time, so as to determine whether data capacity window optimization should be performed. The data fusion synchronization analysis is used to quantify the time alignment of the multi-source sensors in the intelligent connected platform in processing obstacle data. The data capacity window optimization is used to improve the intelligent connected platform's ability to fuse and synchronize obstacle data. The collision risk warning and analysis module is used to perform collision risk assessment and analysis on the obstacle movement trajectory of the intelligent connected platform based on the acquired collision risk prediction data, so as to determine whether to perform filtering and caching optimization. The collision risk assessment and analysis is used to quantify the accuracy of the intelligent connected platform in predicting the collision risk between the intelligent connected vehicle and the obstacle. The filtering and caching optimization is used to improve the accuracy of collision risk prediction and the reliability of collision risk warning. The specific steps for locating, identifying, and analyzing obstacles in the intelligent connected platform's obstacle perception process based on the acquired obstacle perception data are as follows: At the end of the obstacle perception period, the acquired obstacle perception data is compared with the preset obstacle perception data in the database. At the same time, the difference comparison results are corrected by combining the obstacle perception data correction value to obtain the obstacle perception data score. The data is then coupled to obtain the obstacle perception impact index. The obstacle perception data includes antenna array response time, multipath interference ranging time, and radar reception response time. The antenna array response time reflects the time required for the antenna array to respond from receiving a signal during obstacle perception. The multipath interference ranging time reflects the time required for the intelligent connected platform to complete accurate ranging operations in a multipath interference environment. The radar reception response time reflects the time required for the millimeter-wave radar to complete signal processing and provide an effective response after receiving a signal reflected from an obstacle. The obstacle perception impact index is used to quantify the degree of influence of obstacle perception data on the perception capability of the intelligent connected platform.

2. The complex urban environment perception and collision warning system based on intelligent connected vehicles as described in claim 1, characterized in that, The specific steps for determining whether to perform radar wave adjustment and optimization are as follows: If the acquired obstacle perception impact index is greater than the preset obstacle perception impact index in the database, it is recorded as an obstacle perception failure and radar wave adjustment optimization is performed. The radar wave adjustment optimization means improving the intelligent connected platform's obstacle perception response by adjusting the radar wave transmission power and beamwidth. If the obtained obstacle perception impact index is not greater than the preset obstacle perception impact index in the database, it is recorded as qualified for obstacle perception and data fusion and synchronous analysis are performed.

3. The complex urban environment perception and collision warning system based on intelligent connected vehicles as described in claim 2, characterized in that, The radar wave modulation and optimization process involves the following steps: The radar wave transmission power adjustment value obtained by mapping the obtained obstacle perception impact index deviation in the database is used to quantify the specific degree to which the radar wave intensity needs to be increased, so as to reduce the probability of obstacle perception error caused by signal interference. After the radar wave transmission power is adjusted, a transmission power verification command is sent to prompt the preset personnel to conduct a positioning and identification verification test based on the radar wave transmission power after the adjustment. After the positioning and recognition verification test is completed, the obstacle perception process is repeated. If the newly acquired obstacle perception impact index is greater than the preset obstacle perception impact index, an obstacle perception warning is issued; otherwise, the radar wave transmission power is adjusted and data fusion and synchronous analysis are performed.

4. The complex urban environment perception and collision warning system based on intelligent connected vehicles as described in claim 3, characterized in that, The radar wave conditioning optimization also includes beamwidth optimization, with the specific steps as follows: Based on the deviation of the obstacle perception impact index that is re-acquired at the end of the re-performed obstacle perception process, a beamwidth adjustment value is mapped in the database to quantify the degree to which the beamwidth needs to be reduced, so as to reduce signal scrambling interference caused by dense obstacles. After beamwidth adjustment, if the newly acquired obstacle perception impact index is not greater than the preset obstacle perception impact index in the database, the radar wave adjustment and optimization are completed and data fusion and synchronous analysis are performed; otherwise, obstacle perception risk warning is issued.

5. The complex urban environment perception and collision warning system based on intelligent connected vehicles as described in claim 1, characterized in that, The specific steps for determining whether to perform data capacity window optimization are as follows: The acquired obstacle synchronous recognition time is compared with the preset obstacle synchronous recognition time in the database. The obstacle synchronous recognition time represents the time from when the intelligent connected platform receives environmental perception data to when it completes obstacle localization. If the acquired obstacle synchronous recognition time is longer than the preset obstacle synchronous recognition time in the database, it is recorded as unqualified synchronous recognition and data compression optimization is performed. The data compression optimization means improving the intelligent connected platform's response to obstacle location recognition by adjusting the data compression ratio. If the obtained obstacle synchronous recognition time is not greater than the preset obstacle synchronous recognition time in the database, it is recorded as qualified for synchronous recognition and a collision risk assessment analysis is performed.

6. The complex urban environment perception and collision warning system based on intelligent connected vehicles as described in claim 5, characterized in that, The data compression optimization process is as follows: Based on the obtained obstacle synchronous recognition time deviation, the corresponding data compression ratio adjustment value is mapped in the database to reduce the recognition deviation caused by the blurring of environmental perception data; After adjusting the data compression ratio, if the newly acquired obstacle synchronous recognition time is greater than the preset obstacle synchronous recognition time in the database, then capacity window optimization is performed; otherwise, data capacity window optimization is completed and collision risk assessment analysis is performed.

7. The complex urban environment perception and collision warning system based on intelligent connected vehicles as described in claim 6, characterized in that, The capacity window optimization specifically includes: The obstacle synchronous recognition time deviation re-acquired after data compression optimization is mapped in the database to obtain the buffer capacity adjustment value, which is used to quantify the degree to which the buffer capacity needs to be increased, so as to improve the integrity of the intelligent connected platform's recognition data. After the buffer capacity adjustment is completed, if the obstacle synchronization recognition time is not greater than the preset obstacle synchronization recognition time in the database, the data capacity window optimization is completed. Otherwise, the corresponding parallel processing window adjustment value is mapped in the database based on the obstacle synchronization recognition time deviation. This value is used to quantify the degree to which the size of the parallel processing window needs to be increased, thereby reducing task queuing delay and increasing the processing rate of multi-parallel threads. After the parallel processing window is adjusted, if the newly acquired obstacle synchronous recognition time is greater than the preset obstacle synchronous recognition time in the database, a data fusion warning will be issued; otherwise, the data capacity window will be optimized and a collision risk assessment analysis will be performed.

8. The complex urban environment perception and collision warning system based on intelligent connected vehicles as described in claim 1, characterized in that, The step of conducting a collision risk assessment and analysis on the obstacle movement trajectory of the intelligent connected platform based on the acquired collision risk prediction data specifically includes: The obtained trajectory prediction delay time is compared with the preset trajectory prediction delay time in the database to obtain the trajectory prediction delay time score. At the same time, the obtained ranging delay time is compared with the preset ranging delay time in the database to obtain the ranging delay time score. The obtained ranging delay time score and trajectory prediction delay time score are coupled to obtain a collision risk prediction index. The collision risk prediction index is used to quantify the impact of collision risk prediction data on the risk warning capability of the intelligent connected platform. The steps for determining whether to perform filter caching optimization are as follows: If the obtained collision risk prediction index is greater than the preset collision risk prediction index in the database, it is determined that there is a collision risk and a collision risk warning is issued; otherwise, the mean filtering intensity is optimized.

9. The complex urban environment perception and collision warning system based on intelligent connected vehicles as described in claim 8, characterized in that, The optimization of the mean filtering strength is specifically as follows: The mean filter strength adjustment value is obtained by mapping the obtained collision risk prediction index deviation into the database. This value is used to quantify the degree to which the Kalman filter strength needs to be increased in order to reduce the deviation of the target prediction due to noise interference. After the filter strength is adjusted, if the newly acquired collision risk prediction index is greater than the database's preset collision risk prediction index, a collision risk warning will be issued; otherwise, the cache length will be adjusted. The specific process of adjusting the cache length is as follows: Based on the collision risk prediction index deviation re-acquired after the filter strength adjustment, the historical data cache length adjustment value is mapped in the database to quantify the current need to increase the historical cache length, so as to reduce the deviation and misjudgment caused by short-term data fluctuations; after the cache length is adjusted, if the re-acquired collision risk prediction index is greater than the preset collision risk prediction index in the database, a collision risk warning is issued; otherwise, a command indicating that there is currently no collision risk is sent.

Images