Reversing control system and method for electric automobile

Abstract

According to the reversing control system and method for the electric automobile, image information and distance information around the electric automobile are collected in real time, compared with a traditional single reversing radar or reversing image, multi-dimensional environment perception is achieved, hidden dangers such as irregular objects and low obstacles which are difficult to perceive by a traditional system can be effectively recognized, and the reversing control system and method for the electric automobile are suitable for popularization and application. According to the invention, the adjustment direction and the adjustment speed of the driver in the reversing process are determined on the basis of the image information and the distance information, and the voice prompt information is generated on the basis of the adjustment direction and the adjustment speed, so that the driver can be helped to quickly and accurately complete the reversing operation, repeated adjustment caused by misjudgment is avoided, and the driving safety is improved. After the driver selects the automatic parking function, the reversing control information is generated based on the image information and the distance information, and the driver only needs to start the automatic parking function, automatically plan the optimal reversing path and accurately control the vehicle to complete parking without manual operation, so that efficient reversing is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile control, and in particular to a reverse control system and method for an electric vehicle. Background Art

[0002] With the popularity of electric vehicles, reversing safety and convenience have become important issues of concern to users. Most existing electric vehicle reversing systems rely solely on reversing images or reversing radars. The former cannot clearly present the rear environment when the light is insufficient or the image is dirty, making it difficult for the driver to accurately judge obstacles; the latter can only indicate the distance through sound, but cannot provide the specific location and shape of the obstacle. Some systems with automatic reversing functions have problems with low recognition accuracy and slow response speed when identifying complex obstacles, making it difficult to adjust the reversing speed and direction in real time, which can easily lead to collision accidents. In addition, traditional reversing systems lack an integrated solution that organically integrates obstacle recognition, speed adjustment, direction control, and automatic parking functions, resulting in heavy operating burdens for drivers, low reversing efficiency, and major safety hazards. Summary of the Invention

[0003] The present invention provides an electric vehicle reversing control system and method, which are used to solve the problems raised in the background technology.

[0004] An electric vehicle reversing control system, comprising:

[0005] An information acquisition module is used to collect image information and distance information around the electric vehicle in real time;

[0006] A reversing adjustment module is used to determine the driver's adjustment direction and speed during the reversing process based on image information and distance information, and generate a voice reminder message based on the adjustment direction and speed;

[0007] The automatic reversing module is used to generate reversing control information based on image information and distance information after the driver selects the automatic parking function.

[0008] Preferably, the information collection module includes:

[0009] An acquisition and recognition unit is used to acquire environmental images of the rear and left and right sides of the electric vehicle, and to perform obstacle recognition based on the environmental images to obtain image information;

[0010] The distance acquisition unit is used to install distance sensors at different positions of the electric vehicle to collect the distance between the electric vehicle and surrounding obstacles and obtain distance information.

[0011] Preferably, the acquisition and identification unit includes:

[0012] a compensation unit configured to extract features from the environment image to obtain a background image and an obstacle image, predict an environmental index on the background image to obtain an index prediction value, determine a lighting compensation factor, a weather compensation factor, and a scene compensation factor based on a difference between a preset index value and the index prediction value, and compensate the obstacle image based on the lighting compensation factor, the weather compensation factor, and the scene compensation factor to obtain a compensated obstacle image;

[0013] a segmentation unit, configured to perform semantic segmentation on the compensated obstacle image based on a semantic segmentation model to obtain a semantic segmentation result, and perform instance segmentation on the compensated obstacle image based on an instance segmentation model to obtain an instance segmentation result;

[0014] a result determination unit, configured to obtain a common recognition result of the semantic segmentation result and the instance segmentation result for the same feature of the obstacle and a single recognition result for the individual feature of the obstacle, and use the single recognition result as a target recognition result for the individual feature of the obstacle;

[0015] A judgment processing unit, configured to judge whether the difference between the semantic segmentation result and the common recognition result in the instance segmentation result is within a preset range;

[0016] If so, the target recognition result of the same feature of the obstacle is obtained based on the instance segmentation result;

[0017] Otherwise, a first weight is determined based on the association between the target recognition results of the same feature and the individual features of the obstacle and the current recognition result, a second weight is determined based on the dependency of the current recognition result on semantic segmentation, and a third weight is determined based on the dependency of the current recognition result on instance segmentation. The current recognition result is weightedly calculated based on the first weight, the second weight, and the third weight to obtain a target recognition result.

[0018] The information extraction unit is used to extract information based on all target recognition results to obtain image information.

[0019] Preferably, in the judgment processing unit, a weighted calculation is performed on the current recognition result based on the first weight, the second weight and the third weight to obtain a target recognition result, specifically:

[0020] Performing weighted fusion on the semantic segmentation result and the instance segmentation result of the current recognition result based on the second weight and the third weight to obtain a fused recognition result;

[0021] Based on the first weight and in combination with surrounding features, a correction parameter is determined, and the fusion recognition result is corrected based on the correction parameter to obtain a target recognition result.

[0022] Preferably, the reversing adjustment module includes:

[0023] an information determination unit, configured to determine, based on the image information and the distance information, the distance information between the driver and the obstacle during the reversing process, as well as the parking space position and direction information;

[0024] an adjustment determination unit, configured to determine the driver's adjustment direction during the reversing process based on the distance information to the obstacle and the parking space position and direction information, determine an initial adjustment speed based on the difference between the distance information to the obstacle and the standard distance information, and correct the initial adjustment speed based on the steering angle of the adjustment direction to obtain an adjustment speed;

[0025] The reminder determination unit is used to generate voice reminder information based on the difference between the adjustment direction and the adjustment speed and the actual driving situation of the driver.

[0026] Preferably, the reminder determination unit includes:

[0027] a direction reminder unit, configured to determine a direction difference between the driver's actual driving situation and the adjusted direction, and generate a direction deviation reminder and driving direction guidance information based on the direction difference when the direction difference is greater than a preset direction difference, and generate a voice reminder information based on the direction deviation reminder and driving direction guidance information;

[0028] The speed reminder unit is used to determine the speed difference between the driver's actual driving situation and the adjusted speed. When the speed difference is greater than the preset speed difference, a speed deviation reminder and driving speed guidance information are generated based on the speed difference, and a voice reminder information is generated based on the speed deviation reminder and driving speed guidance information.

[0029] Preferably, the automatic reversing module includes:

[0030] An acquisition unit, configured to acquire real-time image information and real-time distance information after the driver selects the automatic parking function, and to acquire initial position information, size information, and parking space information of the electric vehicle;

[0031] a determination unit configured to obtain an initial reversing path and corresponding control parameters based on the real-time image information, the real-time distance information, the initial position information, the size information, and the parking space information of the electric vehicle in combination with a preset reversing control planning model;

[0032] a correction unit, configured to correct the initial reversing path and control parameters based on the dynamic characteristics of the obstacle, the path length, and the control smoothness characteristics, to obtain a target reversing path and target control parameters;

[0033] A monitoring unit, configured to reverse the electric vehicle into a garage according to the target reversing path and target control parameters, and obtain real-time monitoring data;

[0034] The adjustment unit is used to determine and adjust the real-time status of the vehicle based on real-time monitoring data until parking is completed.

[0035] Preferably, the correction unit includes:

[0036] a dynamic area determination unit, configured to predict dynamic features of an obstacle based on real-time image information and real-time distance information, and set a dynamic safety area on the initial reversing path based on the dynamic features of the obstacle;

[0037] a length evaluation unit, configured to discretize the initial reversing path to obtain a plurality of path points, re-search the path points within the dynamic safety area, determine a comprehensive evaluation value of a new path determined by the re-search based on the path length, path turning curvature, and dynamic obstacle risk value, and optimize the initial reversing path based on the comprehensive evaluation value to obtain a plurality of intermediate reversing paths having path lengths within a preset range;

[0038] a smoothness evaluation unit, configured to set a motor response delay constraint and a tire friction limit constraint based on information of the electric vehicle, establish a steering wheel angle change rate, a speed fluctuation rate, and a path tracking error rate as smoothness evaluation indicators, evaluate the plurality of intermediate reversing paths, obtain smoothness evaluation values, and select the path with the largest smoothness evaluation value as the target reversing path;

[0039] The parameter determination unit is used to segment the target reversing path based on the dynamic risk of obstacles and the difficulty of steering wheel operation, obtain the control smoothness of each path segment, determine the initial acceleration of each path segment based on the control smoothness, obtain an acceleration sequence, and perform difference smoothing on the acceleration sequence to obtain a target acceleration sequence. Based on the target acceleration sequence, the target control parameter is obtained in combination with the steering wheel angle.

[0040] A reverse control method for an electric vehicle, comprising:

[0041] S1: Real-time collection of image and distance information around the electric vehicle;

[0042] S2: Determine the driver's adjustment direction and adjustment speed during the reversing process based on the image information and distance information, and generate a voice reminder message based on the adjustment direction and adjustment speed;

[0043] S3: After the driver selects the automatic parking function, reversing control information is generated based on the image information and distance information.

[0044] Preferably, in S3, after the driver selects the automatic parking function, generating reversing control information based on the image information and distance information includes:

[0045] After the driver selects the automatic parking function, real-time image information and real-time distance information are obtained, as well as the initial position information, size information and parking space information of the electric vehicle;

[0046] Based on the real-time image information, real-time distance information, initial position information, size information and parking space information of the electric vehicle, combined with a preset reversing control planning model, an initial reversing path and its corresponding control parameters are obtained;

[0047] Based on the dynamic characteristics of the obstacle, the path length and the control smoothness characteristics, the initial reversing path and the control parameters are modified to obtain the target reversing path and the target control parameters;

[0048] Reversing the electric vehicle into a garage according to the target reversing path and target control parameters, and acquiring real-time monitoring data;

[0049] The real-time status of the vehicle is determined and adjusted based on real-time monitoring data until parking is completed.

[0050] Compared with the prior art, the present invention has achieved the following beneficial effects:

[0051] By collecting image information and distance information around the electric vehicle in real time, multi-dimensional environmental perception is achieved compared to traditional single reversing radar or reversing image. The two complement each other and can effectively identify hidden dangers such as irregular-shaped objects and low obstacles that are difficult for traditional systems to detect, greatly reducing the probability of reversing collision accidents and providing solid protection for driving safety. By determining the driver's adjustment direction and speed during reversing based on image information and distance information, and generating voice reminder information based on the adjustment direction and speed, it can help the driver complete the reversing operation quickly and accurately, and avoid repeated adjustments due to misjudgment. After the driver selects the automatic parking function, the reversing control information is generated based on image information and distance information. The driver does not need to manually operate the steering wheel, accelerator and brake. He only needs to start the automatic parking function, automatically plan the optimal reversing path, and accurately control the vehicle to complete parking, effectively reducing the driver's operating burden and psychological pressure, and achieving efficient reversing.

[0052] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or will be understood by practicing the present invention. The purpose and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in this application document.

[0053] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0055] Figure 1 This is a structural diagram of an electric vehicle reversing control system according to an embodiment of the present invention;

[0056] Figure 2 is a structural diagram of the information acquisition module in an embodiment of the present invention;

[0057] Figure 3 The figure is a flow chart of a reversing control method for an electric vehicle according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0059] Example 1:

[0060] The embodiment of the present invention provides an electric vehicle reversing control system, such as Figure 1 Shown, including:

[0061] An information acquisition module is used to collect image information and distance information around the electric vehicle in real time;

[0062] A reversing adjustment module is used to determine the driver's adjustment direction and speed during the reversing process based on image information and distance information, and generate a voice reminder message based on the adjustment direction and speed;

[0063] The automatic reversing module is used to generate reversing control information based on image information and distance information after the driver selects the automatic parking function.

[0064] In this embodiment, the image information can clearly present the shape, size and specific location of the obstacle.

[0065] In this embodiment, the distance information can accurately measure the distance between the vehicle and the obstacle.

[0066] The beneficial effects of the above design scheme are: by collecting image information and distance information around the electric vehicle in real time, compared with the traditional single reversing radar or reversing image, multi-dimensional environmental perception is achieved. The two complement each other and can effectively identify hidden dangers such as irregular-shaped objects and low obstacles that are difficult for traditional systems to detect, greatly reducing the probability of reversing collision accidents and providing solid protection for driving safety. By determining the driver's adjustment direction and adjustment speed during the reversing process based on image information and distance information, and generating voice reminder information based on the adjustment direction and adjustment speed, it can help the driver complete the reversing operation quickly and accurately, avoiding repeated adjustments due to misjudgment. After the driver selects the automatic parking function, the reversing control information is generated based on the image information and distance information. The driver does not need to manually operate the steering wheel, accelerator and brake, but only needs to start the automatic parking function, automatically plan the optimal reversing path, and accurately control the vehicle to complete parking, effectively reducing the driver's operating burden and psychological pressure, and achieving efficient reversing.

[0067] Example 2:

[0068] Based on Example 1, the present invention provides an electric vehicle reversing control system. Figure 2 As shown, the information collection module includes:

[0069] An acquisition and recognition unit is used to acquire environmental images of the rear and left and right sides of the electric vehicle, and to perform obstacle recognition based on the environmental images to obtain image information;

[0070] The distance acquisition unit is used to install distance sensors at different positions of the electric vehicle to collect the distance between the electric vehicle and surrounding obstacles and obtain distance information.

[0071] In this embodiment, the obstacle information includes the location, shape, and type of the obstacle.

[0072] The beneficial effect of the above design scheme is: by collecting image information and distance information around the electric vehicle in real time, compared with the traditional single reversing radar or reversing image, it realizes multi-dimensional environmental perception. The two complement each other and can effectively identify irregular-shaped objects, low obstacles and other hidden dangers that are difficult for traditional systems to detect, greatly reducing the probability of reversing collision accidents and providing solid protection for driving safety.

[0073] Example 3:

[0074] Based on Example 2, an embodiment of the present invention provides an electric vehicle reversing control system, characterized in that the acquisition and identification unit includes:

[0075] a compensation unit configured to extract features from the environment image to obtain a background image and an obstacle image, predict an environmental index on the background image to obtain an index prediction value, determine a lighting compensation factor, a weather compensation factor, and a scene compensation factor based on a difference between a preset index value and the index prediction value, and compensate the obstacle image based on the lighting compensation factor, the weather compensation factor, and the scene compensation factor to obtain a compensated obstacle image;

[0076] a segmentation unit, configured to perform semantic segmentation on the compensated obstacle image based on a semantic segmentation model to obtain a semantic segmentation result, and perform instance segmentation on the compensated obstacle image based on an instance segmentation model to obtain an instance segmentation result;

[0077] a result determination unit, configured to obtain a common recognition result of the semantic segmentation result and the instance segmentation result for the same feature of the obstacle and a single recognition result for the individual feature of the obstacle, and use the single recognition result as a target recognition result for the individual feature of the obstacle;

[0078] A judgment processing unit, configured to judge whether the difference between the semantic segmentation result and the common recognition result in the instance segmentation result is within a preset range;

[0079] If so, the target recognition result of the same feature of the obstacle is obtained based on the instance segmentation result;

[0080] Otherwise, a first weight is determined based on the association between the target recognition results of the same feature and the individual features of the obstacle and the current recognition result, a second weight is determined based on the dependency of the current recognition result on semantic segmentation, and a third weight is determined based on the dependency of the current recognition result on instance segmentation. The current recognition result is weightedly calculated based on the first weight, the second weight, and the third weight to obtain a target recognition result.

[0081] The information extraction unit is used to extract information based on all target recognition results to obtain image information.

[0082] In this embodiment, the environmental indicators include light intensity, weather conditions, scene complexity, etc.

[0083] In this embodiment, both the semantic segmentation model and the instance segmentation model are obtained based on historical vehicle driving data and combined with deep learning training.

[0084] In this embodiment, the practical application of the compensation factor is, for example, that at night or in strong light environments, the illumination compensation factor can enhance the contrast of obstacles and increase the image recognition accuracy by 30%; in heavy rain or dense fog, the weather compensation factor can effectively suppress noise interference and reduce the false detection rate to below 5%.

[0085] In this embodiment, the common recognition result of the same feature of the obstacle means that both the semantic segmentation result and the instance segmentation result recognize the features of the area and obtain the recognition results. The common recognition result includes the recognition results of the semantic segmentation result and the instance segmentation result for the area. Correspondingly, the single recognition result of the individual feature of the obstacle means that the recognition of the area is only one of the semantic segmentation result and the instance segmentation result, and the other model does not realize the recognition of the area.

[0086] In this embodiment, the semantic segmentation model excels at classification, such as distinguishing between vehicles, pedestrians, and obstacle types, and the instance segmentation model excels at precise positioning, such as bounding boxes and pixel-level masks.

[0087] The beneficial effects of the above design scheme are: by extracting the background image to predict and compensate for environmental indicators, the quality of the compensated obstacle image is guaranteed, and high-quality image features are provided for the acquisition of image information. The compensated obstacle image is analyzed separately based on the semantic segmentation model and the instance segmentation model to obtain recognition results, the recognition results of the individual areas are directly determined, and the recognition results of the common recognition areas are judged and analyzed. Based on the association between the target recognition results of the same features and individual features of the obstacle and the current recognition result, a first weight is determined, a second weight is determined based on the dependence of the current recognition result on semantic segmentation, and a third weight is determined based on the dependence of the current recognition result on instance segmentation. The current recognition result is weightedly calculated based on the first weight, the second weight and the third weight to obtain the target recognition result, thereby ensuring the accuracy of the recognition result and ultimately ensuring the accuracy of the obtained obstacle information, providing accurate information for the reversing control of the electric vehicle.

[0088] Example 4:

[0089] Based on Example 3, an embodiment of the present invention provides an electric vehicle reversing control system. In the judgment processing unit, a weighted calculation is performed on the current recognition result based on the first weight, the second weight, and the third weight to obtain a target recognition result, specifically:

[0090] Performing weighted fusion on the semantic segmentation result and the instance segmentation result of the current recognition result based on the second weight and the third weight to obtain a fused recognition result;

[0091] Based on the first weight and in combination with surrounding features, a correction parameter is determined, and the fusion recognition result is corrected based on the correction parameter to obtain a target recognition result.

[0092] In this embodiment, the first weight is determined by the degree of influence of surrounding features on the current recognition result.

[0093] The beneficial effect of the above design scheme is: by weightedly fusing the semantic segmentation result and the instance segmentation result of the current recognition result based on the second weight and the third weight, a fused recognition result is obtained; based on the first weight and combined with the surrounding features, a correction parameter is determined; based on the correction parameter, the fused recognition result is corrected to obtain a target recognition result; the result is determined from three aspects: the fusion of the two recognition results and the correction of the surrounding features, to ensure the accuracy of the target recognition result.

[0094] Example 5:

[0095] Based on Example 1, this embodiment of the present invention provides an electric vehicle reversing control system, wherein the reversing adjustment module includes:

[0096] an information determination unit, configured to determine, based on the image information and the distance information, the distance information between the driver and the obstacle during the reversing process, as well as the parking space position and direction information;

[0097] an adjustment determination unit, configured to determine the driver's adjustment direction during the reversing process based on the distance information to the obstacle and the parking space position and direction information, determine an initial adjustment speed based on the difference between the distance information to the obstacle and the standard distance information, and correct the initial adjustment speed based on the steering angle of the adjustment direction to obtain an adjustment speed;

[0098] The reminder determination unit is used to generate voice reminder information based on the difference between the adjustment direction and the adjustment speed and the actual driving situation of the driver.

[0099] The beneficial effect of the above design scheme is: by determining the driver's adjustment direction and adjustment speed during the reversing process based on image information and distance information, and generating voice reminder information based on the adjustment direction and adjustment speed, it can help the driver complete the reversing operation quickly and accurately, and avoid repeated adjustments due to misjudgment.

[0100] Example 6:

[0101] Based on Example 5, an embodiment of the present invention provides an electric vehicle reversing control system, wherein the reminder determination unit includes:

[0102] a direction reminder unit, configured to determine a direction difference between the driver's actual driving situation and the adjusted direction, and generate a direction deviation reminder and driving direction guidance information based on the direction difference when the direction difference is greater than a preset direction difference, and generate a voice reminder information based on the direction deviation reminder and driving direction guidance information;

[0103] The speed reminder unit is used to determine the speed difference between the driver's actual driving situation and the adjusted speed. When the speed difference is greater than the preset speed difference, a speed deviation reminder and driving speed guidance information are generated based on the speed difference, and a voice reminder information is generated based on the speed deviation reminder and driving speed guidance information.

[0104] The beneficial effects of this design are: through precise difference detection, intelligent graded warnings and personalized interaction, it not only improves reversing safety, but also significantly improves the user experience, helping drivers complete reversing operations quickly and accurately, and avoiding repeated adjustments caused by misjudgment.

[0105] Example 7:

[0106] Based on Example 1, this embodiment of the present invention provides an electric vehicle reversing control system, wherein the automatic reversing module includes:

[0107] An acquisition unit, configured to acquire real-time image information and real-time distance information after the driver selects the automatic parking function, and to acquire initial position information, size information, and parking space information of the electric vehicle;

[0108] a determination unit configured to obtain an initial reversing path and corresponding control parameters based on the real-time image information, the real-time distance information, the initial position information, the size information, and the parking space information of the electric vehicle in combination with a preset reversing control planning model;

[0109] a correction unit, configured to correct the initial reversing path and control parameters based on the dynamic characteristics of the obstacle, the path length, and the control smoothness characteristics, to obtain a target reversing path and target control parameters;

[0110] A monitoring unit, configured to reverse the electric vehicle into a garage according to the target reversing path and target control parameters, and obtain real-time monitoring data;

[0111] The adjustment unit is used to determine and adjust the real-time status of the vehicle based on real-time monitoring data until parking is completed.

[0112] The beneficial effects of the above design scheme are as follows: after the driver selects the automatic parking function, real-time image information and real-time distance information are obtained, and the initial position information, size information and parking space information of the electric vehicle are obtained; based on the real-time image information, real-time distance information, the initial position information, size information and parking space information of the electric vehicle, combined with a preset reversing control planning model, an initial reversing path and its corresponding control parameters are obtained, thereby achieving accurate planning of the initial reversing path; based on the dynamic characteristics of obstacles, path length and control smoothness characteristics, the initial reversing path and control parameters are corrected to obtain a target reversing path and target control parameters, thereby ensuring the practicality, control smoothness and comfort of the target reversing path; the electric vehicle is reversed into the parking space according to the target reversing path and target control parameters, and real-time monitoring data is obtained; the real-time status of the vehicle is determined and adjusted based on the real-time monitoring data until parking is completed, thereby achieving real-time and flexible changes during the parking process and ensuring the safety of parking.

[0113] Example 8:

[0114] Based on Example 7, this embodiment of the present invention provides an electric vehicle reversing control system, wherein the correction unit includes:

[0115] a dynamic area determination unit, configured to predict dynamic features of an obstacle based on real-time image information and real-time distance information, and set a dynamic safety area on the initial reversing path based on the dynamic features of the obstacle;

[0116] a length evaluation unit, configured to discretize the initial reversing path to obtain a plurality of path points, re-search the path points within the dynamic safety area, determine a comprehensive evaluation value of a new path determined by the re-search based on the path length, path turning curvature, and dynamic obstacle risk value, and optimize the initial reversing path based on the comprehensive evaluation value to obtain a plurality of intermediate reversing paths having path lengths within a preset range;

[0117] a smoothness evaluation unit, configured to set a motor response delay constraint and a tire friction limit constraint based on information of the electric vehicle, establish a steering wheel angle change rate, a speed fluctuation rate, and a path tracking error rate as smoothness evaluation indicators, evaluate the plurality of intermediate reversing paths, obtain smoothness evaluation values, and select the path with the largest smoothness evaluation value as the target reversing path;

[0118] The parameter determination unit is used to segment the target reversing path based on the dynamic risk of obstacles and the difficulty of steering wheel operation, obtain the control smoothness of each path segment, determine the initial acceleration of each path segment based on the control smoothness, obtain an acceleration sequence, and perform difference smoothing on the acceleration sequence to obtain a target acceleration sequence. Based on the target acceleration sequence, the target control parameter is obtained in combination with the steering wheel angle.

[0119] In this embodiment, a larger comprehensive evaluation value indicates a greater probability of being selected as the intermediate reversing path. A larger comprehensive evaluation value indicates that the turning point of the initial reversing path is smoothed, the invalid distance is shortened, and the optimal path length is ensured.

[0120] The beneficial effects of the above design scheme are: through dynamic safety area division and path point discretization processing, it can respond to changes in moving obstacles in real time, reduce collision risks, and ensure the optimal balance between path length and steering curvature. It introduces motor response delay and tire friction constraints, evaluates path smoothness through indicators such as steering wheel angle change rate, significantly reduces sharp turns and speed mutations, and improves driving comfort. It optimizes acceleration sequences based on dynamic risk segmentation and combines steering wheel angle parameter matching to achieve smooth transitions under different road conditions and reduce system energy consumption. It predicts the dynamic characteristics of obstacles through the fusion of real-time images and distance information, enabling the system to adapt to complex parking environments and improve the success rate and safety of automatic parking.

[0121] Example 9:

[0122] The embodiment of the present invention provides a method for controlling the reverse movement of an electric vehicle. Figure 3 Shown, including:

[0123] S1: Real-time collection of image and distance information around the electric vehicle;

[0124] S2: Determine the driver's adjustment direction and adjustment speed during the reversing process based on the image information and distance information, and generate a voice reminder message based on the adjustment direction and adjustment speed;

[0125] S3: After the driver selects the automatic parking function, reversing control information is generated based on the image information and distance information.

[0126] In this embodiment, the image information can clearly present the shape, size and specific location of the obstacle.

[0127] In this embodiment, the distance information can accurately measure the distance between the vehicle and the obstacle.

[0128] The beneficial effects of the above design scheme are: by collecting image information and distance information around the electric vehicle in real time, compared with the traditional single reversing radar or reversing image, multi-dimensional environmental perception is achieved. The two complement each other and can effectively identify hidden dangers such as irregular-shaped objects and low obstacles that are difficult for traditional systems to detect, greatly reducing the probability of reversing collision accidents and providing solid protection for driving safety. By determining the driver's adjustment direction and adjustment speed during the reversing process based on image information and distance information, and generating voice reminder information based on the adjustment direction and adjustment speed, it can help the driver complete the reversing operation quickly and accurately, avoiding repeated adjustments due to misjudgment. After the driver selects the automatic parking function, the reversing control information is generated based on the image information and distance information. The driver does not need to manually operate the steering wheel, accelerator and brake, but only needs to start the automatic parking function, automatically plan the optimal reversing path, and accurately control the vehicle to complete parking, effectively reducing the driver's operating burden and psychological pressure, and achieving efficient reversing.

[0129] Example 10:

[0130] Based on Example 9, an embodiment of the present invention provides a method for controlling reversing of an electric vehicle. In S3, after the driver selects the automatic parking function, reversing control information is generated based on image information and distance information, including:

[0131] After the driver selects the automatic parking function, real-time image information and real-time distance information are obtained, as well as the initial position information, size information and parking space information of the electric vehicle;

[0132] Based on the real-time image information, real-time distance information, initial position information, size information and parking space information of the electric vehicle, combined with a preset reversing control planning model, an initial reversing path and its corresponding control parameters are obtained;

[0133] Based on the dynamic characteristics of the obstacle, the path length and the control smoothness characteristics, the initial reversing path and the control parameters are modified to obtain the target reversing path and the target control parameters;

[0134] Reversing the electric vehicle into a garage according to the target reversing path and target control parameters, and acquiring real-time monitoring data;

[0135] The real-time status of the vehicle is determined and adjusted based on real-time monitoring data until parking is completed.

[0136] The beneficial effects of the above design scheme are as follows: after the driver selects the automatic parking function, real-time image information and real-time distance information are obtained, and the initial position information, size information and parking space information of the electric vehicle are obtained; based on the real-time image information, real-time distance information, the initial position information, size information and parking space information of the electric vehicle, combined with a preset reversing control planning model, an initial reversing path and its corresponding control parameters are obtained, thereby achieving accurate planning of the initial reversing path; based on the dynamic characteristics of obstacles, path length and control smoothness characteristics, the initial reversing path and control parameters are corrected to obtain a target reversing path and target control parameters, thereby ensuring the practicality, control smoothness and comfort of the target reversing path; the electric vehicle is reversed into the parking space according to the target reversing path and target control parameters, and real-time monitoring data is obtained; the real-time status of the vehicle is determined and adjusted based on the real-time monitoring data until parking is completed, thereby achieving real-time and flexible changes during the parking process and ensuring the safety of parking.

[0137] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of this application document and its equivalents, the present invention is intended to include these modifications and variations.

Claims

1. An electric vehicle reversing control system, characterized in that: include: An information acquisition module is used to collect image information and distance information around the electric vehicle in real time; A reversing adjustment module is used to determine the driver's adjustment direction and speed during the reversing process based on image information and distance information, and generate a voice reminder message based on the adjustment direction and speed; The automatic reversing module is used to generate reversing control information based on image information and distance information after the driver selects the automatic parking function.

2. The electric vehicle reversing control system according to claim 1, characterized in that: The information collection module includes: An acquisition and recognition unit is used to acquire environmental images of the rear and left and right sides of the electric vehicle, and to perform obstacle recognition based on the environmental images to obtain image information; The distance acquisition unit is used to install distance sensors at different positions of the electric vehicle to collect the distance between the electric vehicle and surrounding obstacles and obtain distance information.

3. The electric vehicle reversing control system according to claim 2, characterized in that: The acquisition and identification unit includes: a compensation unit configured to extract features from the environment image to obtain a background image and an obstacle image, predict an environmental index on the background image to obtain an index prediction value, determine a lighting compensation factor, a weather compensation factor, and a scene compensation factor based on a difference between a preset index value and the index prediction value, and compensate the obstacle image based on the lighting compensation factor, the weather compensation factor, and the scene compensation factor to obtain a compensated obstacle image; a segmentation unit, configured to perform semantic segmentation on the compensated obstacle image based on a semantic segmentation model to obtain a semantic segmentation result, and perform instance segmentation on the compensated obstacle image based on an instance segmentation model to obtain an instance segmentation result; a result determination unit, configured to obtain a common recognition result of the semantic segmentation result and the instance segmentation result for the same feature of the obstacle and a single recognition result for the individual feature of the obstacle, and use the single recognition result as a target recognition result for the individual feature of the obstacle; A judgment processing unit, configured to judge whether the difference between the semantic segmentation result and the common recognition result in the instance segmentation result is within a preset range; If so, the target recognition result of the same feature of the obstacle is obtained based on the instance segmentation result; Otherwise, a first weight is determined based on the association between the target recognition results of the same feature and the individual features of the obstacle and the current recognition result, a second weight is determined based on the dependency of the current recognition result on semantic segmentation, and a third weight is determined based on the dependency of the current recognition result on instance segmentation. The current recognition result is weightedly calculated based on the first weight, the second weight, and the third weight to obtain a target recognition result. The information extraction unit is used to extract information based on all target recognition results to obtain image information.

4. The electric vehicle reversing control system according to claim 3, characterized in that: In the judgment processing unit, a weighted calculation is performed on the current recognition result based on the first weight, the second weight, and the third weight to obtain a target recognition result, specifically: Performing weighted fusion on the semantic segmentation result and the instance segmentation result of the current recognition result based on the second weight and the third weight to obtain a fused recognition result; Based on the first weight and in combination with surrounding features, a correction parameter is determined, and the fusion recognition result is corrected based on the correction parameter to obtain a target recognition result.

5. The electric vehicle reversing control system according to claim 1, characterized in that: The reversing adjustment module includes: an information determination unit, configured to determine, based on the image information and the distance information, the distance information between the driver and the obstacle during the reversing process, as well as the parking space position and direction information; an adjustment determination unit, configured to determine the driver's adjustment direction during the reversing process based on the distance information to the obstacle and the parking space position and direction information, determine an initial adjustment speed based on the difference between the distance information to the obstacle and the standard distance information, and correct the initial adjustment speed based on the steering angle of the adjustment direction to obtain an adjustment speed; The reminder determination unit is used to generate voice reminder information based on the difference between the adjustment direction and the adjustment speed and the actual driving situation of the driver.

6. The electric vehicle reversing control system according to claim 5, characterized in that: The reminder determination unit includes: a direction reminder unit, configured to determine a direction difference between the driver's actual driving situation and the adjusted direction, and generate a direction deviation reminder and driving direction guidance information based on the direction difference when the direction difference is greater than a preset direction difference, and generate a voice reminder information based on the direction deviation reminder and driving direction guidance information; The speed reminder unit is used to determine the speed difference between the driver's actual driving situation and the adjusted speed. When the speed difference is greater than the preset speed difference, a speed deviation reminder and driving speed guidance information are generated based on the speed difference, and a voice reminder information is generated based on the speed deviation reminder and driving speed guidance information.

7. The electric vehicle reversing control system according to claim 1, characterized in that: The automatic reversing module comprises: An acquisition unit, configured to acquire real-time image information and real-time distance information after the driver selects the automatic parking function, and to acquire initial position information, size information, and parking space information of the electric vehicle; a determination unit configured to obtain an initial reversing path and corresponding control parameters based on the real-time image information, the real-time distance information, the initial position information, the size information, and the parking space information of the electric vehicle in combination with a preset reversing control planning model; a correction unit, configured to correct the initial reversing path and control parameters based on the dynamic characteristics of the obstacle, the path length, and the control smoothness characteristics, to obtain a target reversing path and target control parameters; A monitoring unit, configured to reverse the electric vehicle into a garage according to the target reversing path and target control parameters, and obtain real-time monitoring data; The adjustment unit is used to determine and adjust the real-time status of the vehicle based on real-time monitoring data until parking is completed.

8. The electric vehicle reversing control system according to claim 7, characterized in that: The correction unit includes: a dynamic area determination unit, configured to predict dynamic features of an obstacle based on real-time image information and real-time distance information, and set a dynamic safety area on the initial reversing path based on the dynamic features of the obstacle; a length evaluation unit, configured to discretize the initial reversing path to obtain a plurality of path points, re-search the path points within the dynamic safety area, determine a comprehensive evaluation value of a new path determined by the re-search based on the path length, path turning curvature, and dynamic obstacle risk value, and optimize the initial reversing path based on the comprehensive evaluation value to obtain a plurality of intermediate reversing paths having path lengths within a preset range; a smoothness evaluation unit, configured to set a motor response delay constraint and a tire friction limit constraint based on information of the electric vehicle, establish a steering wheel angle change rate, a speed fluctuation rate, and a path tracking error rate as smoothness evaluation indicators, evaluate the plurality of intermediate reversing paths, obtain smoothness evaluation values, and select the path with the largest smoothness evaluation value as the target reversing path; The parameter determination unit is used to segment the target reversing path based on the dynamic risk of obstacles and the difficulty of steering wheel operation, obtain the control smoothness of each path segment, determine the initial acceleration of each path segment based on the control smoothness, obtain an acceleration sequence, and perform difference smoothing on the acceleration sequence to obtain a target acceleration sequence. Based on the target acceleration sequence, the target control parameter is obtained in combination with the steering wheel angle.

9. A method for controlling the reverse movement of an electric vehicle, used in the reverse movement control system of an electric vehicle as claimed in claim 1, characterized in that: include: S1: Real-time collection of image and distance information around the electric vehicle; S2: Determine the driver's adjustment direction and adjustment speed during the reversing process based on the image information and distance information, and generate a voice reminder message based on the adjustment direction and adjustment speed; S3: After the driver selects the automatic parking function, reversing control information is generated based on the image information and distance information.

10. The electric vehicle reversing control method according to claim 9, characterized in that: In S3, after the driver selects the automatic parking function, reversing control information is generated based on the image information and the distance information, including: After the driver selects the automatic parking function, real-time image information and real-time distance information are obtained, as well as the initial position information, size information and parking space information of the electric vehicle; Based on the real-time image information, real-time distance information, initial position information, size information and parking space information of the electric vehicle, combined with a preset reversing control planning model, an initial reversing path and its corresponding control parameters are obtained; Based on the dynamic characteristics of the obstacle, the path length and the control smoothness characteristics, the initial reversing path and the control parameters are modified to obtain the target reversing path and the target control parameters; Reversing the electric vehicle into a garage according to the target reversing path and target control parameters, and acquiring real-time monitoring data; The real-time status of the vehicle is determined and adjusted based on real-time monitoring data until parking is completed.

Images