Driving assistance method and system for vehicle and vehicle

Abstract

The invention provides a driving assistance method and system for a vehicle and the vehicle, and the driving assistance method comprises the steps: collecting the voice features of a driver when a first vehicle runs, and judging whether the voice features are matched with a preset voice database or not; if yes, acquiring first vehicle environment information of a vehicle-mounted sensor, calculating a driving state parameter based on the first vehicle environment information, and judging whether the first vehicle is in a dangerous driving environment or not according to the driving state parameter; if yes, the braking opportunity and the braking force of the first vehicle are calculated according to the driving state parameters, and a braking system of the first vehicle is controlled to execute active braking according to the braking opportunity and the braking force. The technical problems that an existing driving assistance system depends on sensor data to achieve automatic braking, but in an emergency, voice response often reflects danger faster than sensor data or a manual starting system, but in the prior art, a voice starting system cannot be combined in time, intervention lags, and potential safety hazards exist are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle risk avoidance, and more particularly, to a driving assistance method, system, and vehicle for a vehicle. Background Art

[0002] With the rapid development of the automotive industry, driving assistance systems have played an important role in improving driving safety and reducing the accident rate. Traditional driving assistance systems usually rely on in-vehicle sensors (such as radar, cameras, ultrasonic sensors, etc.) to obtain information about the vehicle's surrounding environment, and achieve functions such as automatic braking and lane keeping by analyzing parameters such as relative speed and distance.

[0003] However, there are at least the following problems in the related art: The existing driving assistance system relies on sensor data to achieve automatic braking. However, in an emergency, the voice reaction is often faster than the sensor data or manually turning on the system in reflecting danger, but the existing technology cannot combine the voice to start the system in time, resulting in a lag in intervention and potential safety hazards. Summary of the Invention

[0004] The technical problem solved by the present invention is that the existing driving assistance system relies on sensor data to achieve automatic braking. However, in an emergency, the voice reaction is often faster than the sensor data or manually turning on the system in reflecting danger, but the existing technology cannot combine the voice to start the system in time, resulting in a lag in intervention and potential safety hazards.

[0005] To solve the above problems, the present invention provides a driving assistance method for a vehicle. The driving assistance method includes: collecting the voice characteristics of a driver during the driving process of a first vehicle, and determining whether the voice characteristics match a preset voice database; if so, obtaining the vehicle environment information of an in-vehicle sensor, calculating a driving state parameter based on the vehicle environment information, and determining whether the first vehicle is in a dangerous driving environment according to the driving state parameter; if it is determined that the first vehicle is in a dangerous driving environment, calculating the braking timing and braking force of the first vehicle according to the driving state parameter, and controlling the braking system of the first vehicle to perform active braking according to the braking timing and braking force.

[0006] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By collecting the voice characteristics of the driver in real time, the shouts or other abnormal sounds made by the driver due to the critical situation can be collected in time, triggering the braking system to brake in time before an accident occurs, and making adaptive adjustments to the braking timing and braking force according to the distance between the front and rear vehicles and the vehicle state.

[0007] In an example of the present invention, the driving state parameters are used to describe the relative state between the first vehicle and the second vehicle located in front of the first vehicle. The driving state parameters include: relative speed v, acceleration a, and relative distance d. Calculating the braking timing and braking force of the first vehicle according to the driving state parameters includes: calculating the braking time T according to the relative distance d and the relative speed v. When the braking time T is less than the second threshold T th , controlling the braking system to perform an active braking action; calculating the first braking force F according to the relative distance d and the relative speed v b1 ; calculating the smoothing function σ according to the braking time T and the second threshold T th ; calculating the second braking force F according to the first braking force F b1 and the smoothing function σ b2 , controlling the braking system to perform a braking action according to the second braking force F b2 .

[0008] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By calculating the braking time and the first braking force based on the relative distance and speed, and introducing a smoothing function to calculate the second braking force, precise control of the braking timing and force is achieved, improving the smoothness and comfort of the braking action and further enhancing driving safety.

[0009] In an example of the present invention, calculating the second braking force Fb2 according to the first braking force Fb1 and the smoothing function σ includes: the second braking force F b2 = F b1× σ, and the smoothing function σ satisfies the following calculation formula: ; where α is the slope parameter of the smoothing function σ.

[0010] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By constructing a smoothing function, the smoothness of the braking force during braking is further controlled, avoiding danger to the occupants in the first vehicle caused by emergency braking.

[0011] In an example of the present invention, collecting the voice characteristics of the driver during driving and determining whether the voice characteristics match a preset target voice pattern includes: real-time collecting the voice data of the driver through the voice collection device of the first vehicle, performing voice analysis on the collected voice data to extract voice characteristics; using a preset voice database to match the voice characteristics with the voice database to determine whether the voice characteristics match those in the voice database; if not, controlling the voice collection device to continue collecting the voice data of the driver.

[0012] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By collecting the driver's voice data in real time, extracting voice features, and matching them with a preset voice database, an efficient and accurate voice recognition process is realized. Using the voice collection device and the preset database of the first vehicle, it is possible to quickly determine whether the voice features meet the expectations, providing a reliable trigger condition for subsequent driving assistance decisions and enhancing the response speed and intelligence level of the system.

[0013] In an example of the present invention, the vehicle environment information includes target detection data and image data; the vehicle-mounted sensors include a vehicle-mounted radar sensor and a vehicle-mounted camera. Obtaining the vehicle environment information of the vehicle-mounted sensors includes: obtaining target detection data through the vehicle-mounted radar sensor and obtaining image data through the vehicle-mounted camera; calculating a first state parameter based on the target detection data and calculating a second state parameter through the image data; the first state parameter includes a first relative speed v1, a first acceleration a1, and a first relative distance d1, and the second state parameter includes a second relative speed v2, a second acceleration a2, and a second relative distance d2; the first state parameter and the second state parameter are fused by using a weighted average or a Kalman filtering algorithm to obtain a driving state parameter.

[0014] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By simultaneously using a vehicle-mounted radar sensor and a vehicle-mounted camera to obtain target detection data and image data respectively, multi-dimensional vehicle environment information collection is realized. In addition, by calculating a first state parameter and a second state parameter based on the target detection data and the image data respectively, and fusing the two to obtain a driving state parameter including relative speed, acceleration, and relative distance, comprehensive analysis of multi-dimensional data is realized.

[0015] In an example of the present invention, calculating the second state parameter through the image data includes: extracting the pixel height h of the target object in the image data p and the actual height h of the target object r , and calculating the second relative distance d2 through the pixel height h p and the actual height h r . The second relative distance d2 satisfies the following calculation formula: ; where f is a calibration parameter.

[0016] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: By measuring the ratio of the pixel height to the actual height, the relative distance between the first vehicle and the second vehicle can be further obtained, and the calculation process is accurate.

[0017] In an example of the present invention, a danger state parameter S is calculated based on the relative speed v and the relative distance d. The danger state parameter is used to represent the driving state of the first vehicle: wherein, when the value of the danger state parameter S is 1, it is determined that the first vehicle is in a dangerous driving environment; when the value of the danger state parameter S is 0, it is determined that the first vehicle is not in a dangerous driving environment.

[0018] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: when the first vehicle gradually approaches the second vehicle and the relative distance between them is small, it indicates that there is a collision risk between the two. In this case, it is highly accurate to determine that the first vehicle is in a dangerous state and it is not likely to make misjudgments.

[0019] In an example of the present invention, determining whether the first vehicle is in a dangerous driving environment according to the driving state parameters includes: if the first vehicle is not in a dangerous driving environment, controlling the voice acquisition device to continue to acquire the voice data of the driver.

[0020] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: when the first vehicle is not in a dangerous driving environment, the system will continuously acquire voice data and maintain the monitoring ability of the driver's voice. It ensures the real-time response ability to potential dangers and improves the continuity and practicality of the driving assistance function.

[0021] On the other hand, the present invention also provides a driving assistance system for a vehicle. If the driving assistance method in any of the above examples is applied to the driving assistance system, the driving assistance system includes: a voice module, which is used to acquire the voice characteristics of the driver and determine whether the voice characteristics fall into a preset voice database; a judgment module, which is used to judge whether the first vehicle is in a dangerous driving environment; an execution module, which is used to calculate the braking timing and braking force of the first vehicle and control the first vehicle to brake.

[0022] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: it can achieve the technical effects corresponding to any of the above examples, which will not be elaborated here.

[0023] On the other hand, the present invention also provides a vehicle to which the driving assistance method in any of the above examples is applied.

[0024] Compared with the prior art, the technical effects achieved by adopting this technical solution are as follows: it can achieve the technical effects corresponding to any of the above examples, which will not be elaborated here.

[0025] After adopting the technical solution of the present invention, the following technical effects can be achieved: (1)By collecting the voice characteristics of the driver in real time, even if shouts or other abnormal noises made by the driver due to critical situations are collected, the braking system is triggered in time to brake before an accident occurs, and the braking timing and braking force are adaptively adjusted according to the distance between the front and rear vehicles and the vehicle state; (2)By simultaneously using an in-vehicle radar sensor and an in-vehicle camera to respectively obtain target detection data and image data, multi-dimensional vehicle environment information collection is achieved; (3)By calculating the braking time and the first braking force based on the relative distance and speed, and introducing a smoothing function to calculate the second braking force, precise control of the braking timing and force is achieved, the smoothness and comfort of the braking action are improved, and the driving safety is further enhanced. Description of the Drawings

[0026] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts; Figure 1 It is a flowchart of a driving assistance method for a vehicle provided by an embodiment of the present invention; Figure 2 It is a structural schematic block diagram of a driving assistance system for a vehicle provided by an embodiment of the present invention.

[0027] Description of the Reference Numerals: 100, driving assistance system; 101, voice module; 102, judgment module; 103, execution module. Detailed Embodiments

[0028] To make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings.

[0029] See Figure 1 , the present invention provides a driving assistance method for a vehicle, and the driving assistance method includes the following steps: S100: Collect the voice characteristics of the driver during the driving process of the first vehicle, and determine whether the voice characteristics match a preset voice database; S200: If so, obtain the vehicle environment information of the in-vehicle sensor, calculate the driving state parameters based on the vehicle environment information, and determine whether the first vehicle is in a dangerous driving environment according to the driving state parameters; S300: If it is determined that the vehicle is in a dangerous driving environment, calculate the braking timing and braking force of the first vehicle according to the driving state parameters, and control the braking system of the first vehicle to perform active braking according to the braking timing and braking force.

[0030] Preferably, by collecting the voice characteristics of the driver in real time, the shouts or other abnormal noises made by the driver due to critical situations can be collected in time, the braking system can be triggered in time to brake before an accident occurs, and the braking timing and braking force can be adaptively adjusted according to the distance between the front and rear vehicles and the state of the first vehicle.

[0031] Preferably, when collecting voice characteristics, it is not limited to the driver in the driving position, and the voice characteristics of other personnel in the first vehicle are also within the scope of the embodiments of the present application.

[0032] Preferably, in the present application, the voice characteristics can also collect other abnormal noises inside the first vehicle, not limited to the abnormal voices of simple personnel, such as the sound of heavy objects colliding and falling inside the first vehicle or the abnormal noises of the first vehicle itself, etc., which may affect the driver or the running state of the first vehicle.

[0033] Further, S300 includes: S310: The driving state parameters are used to describe the relative state between the first vehicle and the second vehicle located in front of the first vehicle. The driving state parameters include: relative speed v, acceleration a, and relative distance d; calculate the braking time T according to the relative distance d and relative speed v. When the braking time T is less than the second threshold T th control the braking system to perform an active braking action; calculate the first braking force F according to the relative distance d and relative speed v b1 ; calculate the smoothing function σ according to the braking time T and the second threshold T th calculate the second braking force F according to the first braking force F b1 and the smoothing function σ b2 control the braking system to perform a braking action according to the second braking force F b2 .

[0034] Specifically: The braking time T satisfies the following formula 1: Formula 1: ; Further, the first braking force F b1 satisfies the following formula 2: Formula 2: ; where k1 and k2 are weighting coefficients; Further, calculating the second braking force Fb2 according to the first braking force Fb1 and the smoothing function σ includes: The second braking force F b2 = Fb1× σ, a smoothing function σ satisfies the following calculation formula 3: Formula 3: ; where α is the slope parameter of the smoothing function σ.

[0035] Preferably, by calculating the braking time and the first braking force based on the relative distance and speed, and introducing a smoothing function to calculate the second braking force, precise control of the braking timing and force is achieved, improving the smoothness and comfort of the braking action and further enhancing driving safety.

[0036] Preferably, k1 and k2 are calibrated through experiments to determine the braking scheme most suitable for the first vehicle.

[0037] Preferably, in this embodiment, when T is less than the second threshold T th perform an active braking action, and when the relative speed is 0, end the active braking action.

[0038] Furthermore, S100 includes: S110: Real-time collect the voice data of the driver through the voice collection device of the first vehicle, perform voice analysis on the collected voice data to extract voice features; S120: Use a preset voice database to match the voice features with the voice database to determine whether the voice features match those in the voice database.

[0039] S130: If not, control the voice collection device to continue collecting the voice data of the driver.

[0040] Preferably, by real-time collecting the driver's voice data and performing the matching of voice feature extraction with the preset voice database, an efficient and accurate voice recognition process is achieved. Using the voice collection device of the first vehicle and the preset database, it is possible to quickly determine whether the voice features meet the expectations, providing a reliable trigger condition for subsequent driving assistance decisions and enhancing the response speed and intelligence level of the system.

[0041] Preferably, when performing voice analysis, the voice features are divided into semantic features and volume features, which respectively represent the volume of the people inside the first vehicle and the keywords spoken, and the combination of volume and semantics is used to determine whether the collected voice data will affect the driver.

[0042] Preferably, the semantic feature keywords include panicked shouts such as "ahhhhhh".

[0043] Preferably, when matching the voice features with the voice database, an AI database is used for auxiliary analysis to accurately determine the source of the voice features and the scenario when the voice features are generated, and further analyze whether the scenario will cause a vehicle accident.

[0044] Preferably, when the voice feature does not match the preset voice database, the system continuously collects the driver's voice data to ensure the continuity of the voice recognition process.

[0045] Furthermore, the vehicle environment information includes target detection data and image data, and S200 includes: S210: Obtain the target detection data through an in-vehicle radar sensor and obtain the image data through an in-vehicle camera; S220: Calculate the first state parameter based on the target detection data and calculate the second state parameter through the image data; S230: Fuse the first state parameter and the second state parameter by using a weighted average or Kalman filter algorithm to obtain the driving state parameter.

[0046] Furthermore, the first state parameter includes a first relative speed v1, a first acceleration a1, and a first relative distance d1, and the second state parameter includes a second relative speed v2, a second acceleration a2, and a second relative distance d2; Preferably, by simultaneously using an in-vehicle radar sensor and an in-vehicle camera to obtain the target detection data and the image data respectively, multi-dimensional vehicle environment information collection is realized. The radar sensor provides accurate target distance and speed information, and the camera provides rich visual information. The combination of the two enhances the comprehensiveness and accuracy of environmental perception.

[0047] Preferably, if there is no second vehicle ahead but other obstacles, the driving assistance method of the embodiment of the present application is also applicable, and the calculation method of other obstacles can be equivalently replaced with that of the second vehicle.

[0048] Preferably, by calculating the first state parameter and the second state parameter based on the target detection data and the image data respectively, and fusing the two to obtain the driving state parameter including relative speed, acceleration, and relative distance, multi-dimensional data comprehensive analysis is realized. The accuracy and reliability of the driving state parameter are improved, providing more comprehensive data support for the judgment of dangerous driving environments.

[0049] Preferably, the in-vehicle radar sensor measures the target detection data through the Doppler effect, and extracting the target detection data shows that the speed of the vehicle ahead is v t , and the speed of the vehicle itself is v e , then the first relative speed v1 = v t - v e , and then the first acceleration a1 can be calculated.

[0050] Preferably, if there is no vehicle ahead, the first acceleration a1 of the vehicle itself can be directly measured by an accelerometer.

[0051] Preferably, the vehicle-mounted radar sensor can directly obtain the first relative distance d1 between the host vehicle and the target object.

[0052] Preferably, the vehicle-mounted camera can perform image processing on the image data to obtain the second relative velocity v2 between the host vehicle and the target object, and then calculate the first acceleration a2. The image processing method can be the optical flow method or target detection.

[0053] Preferably, the pixel height h of the target object in the image data is extracted p and the actual height h of the target object r , and the second relative distance d2 is calculated through the pixel height h p and the actual height h r . The second relative distance d2 satisfies the following calculation formula 4: Formula 4: ; where f is a calibration parameter, and its specific meaning is the focal length of the vehicle-mounted camera.

[0054] Preferably, the fusion algorithm uses weighted average or Kalman filter algorithm to fuse the first state parameter and the second state parameter.

[0055] Furthermore, S200 further includes: S240: Calculate the dangerous state parameter S through the relative velocity v and the relative distance d. The dangerous state parameter is used to represent the driving state of the first vehicle: where, when the value of the dangerous state parameter S is 1, it is determined that the first vehicle is in a dangerous driving environment; when the value of the dangerous state parameter S is 0, it is determined that the first vehicle is not in a dangerous driving environment.

[0056] Furthermore, the dangerous state parameter S satisfies the following formula 5: Formula 5: ; where D th in the above formula is the first threshold, and the relative velocity v less than 0 indicates that the first vehicle and the second first vehicle are approaching each other.

[0057] Preferably, when the first vehicle gradually approaches the vehicle in front and the relative distance between them is small, it indicates that there is a collision risk between the two. In this case, it is highly accurate to determine that the first vehicle is in a dangerous state and it is not likely to make a misjudgment.

[0058] Preferably, the dangerous judgment method further includes lane departure warning. When it is detected that the traveling direction of the first vehicle has a large deviation, it is determined that the first vehicle is in a dangerous driving environment, and the braking system is controlled to keep driving in the current lane.

[0059] Furthermore, S200 further includes: S250: If the first vehicle is not in a dangerous driving environment, control the voice acquisition device to continue acquiring the driver's voice data.

[0060] Preferably, when the first vehicle is not in a dangerous driving environment, the system will continuously acquire voice data to maintain the monitoring ability of the driver's voice. This ensures the real-time response ability to potential dangers and improves the continuity and practicality of the driving assistance function.

[0061] See Figure 2 , on the other hand, the present invention also provides a driving assistance system 100 for a vehicle. When the driving assistance method in any of the above examples is applied to the driving assistance system 100, the driving assistance system 100 includes: A voice module 101, which is used to obtain the voice characteristics of the driver and determine whether the voice characteristics fall into a preset voice database; A judgment module 102, which is used to judge whether the first vehicle is in a dangerous driving environment: An execution module 103, which is used to calculate the braking timing and braking force of the first vehicle and control the first vehicle to brake.

[0062] On the other hand, the present invention also provides a vehicle to which the driving assistance method in any of the above examples is applied.

[0063] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be subject to the scope defined by the claims.

Claims

1. A driving assistance method for a vehicle, characterized in that, The driving assistance method includes: Collecting the voice characteristics of the driver during the driving process of the first vehicle, and determining whether the voice characteristics match a preset voice database; If so, obtaining the vehicle environment information of the vehicle-mounted sensor, calculating the driving state parameters based on the first vehicle environment information, and determining whether the first vehicle is in a dangerous driving environment according to the driving state parameters; If it is determined that the vehicle is in the dangerous driving environment, calculating the braking timing and braking force of the first vehicle according to the driving state parameters, and controlling the braking system of the first vehicle to perform active braking according to the braking timing and the braking force.

2. The driving assistance method according to claim 1, wherein The driving state parameters are used to describe the relative state between the first vehicle and a second vehicle located in front of the first vehicle, and the driving state parameters include: relative speed v, acceleration a, and relative distance d; Calculating the braking timing and braking force of the first vehicle according to the driving state parameters includes: Calculate the braking time T based on the relative distance d and the relative speed v. When the braking time T is less than the second threshold T th , control the braking system to perform an active braking action; Calculating a first braking force Fb1 according to the relative distance d and the relative speed v; According to the braking time T and the second threshold T th Calculate the smoothing function σ, calculate the second braking force Fb2 according to the first braking force Fb1 and the smoothing function σ, and control the braking system to perform a braking action according to the second braking force Fb2.

3. The driving assistance method according to claim 2, wherein Calculating a second braking force Fb2 according to the first braking force Fb1 and a smoothing function σ includes: The second braking force F b2 = F b1× σ, and the smoothing function σ satisfies the following calculation formula: ； where α is the slope parameter of the smoothing function σ.

4. The driving assistance method according to claim 1, wherein Collecting the voice characteristics of the driver during the driving process of the first vehicle and determining whether the voice characteristics match a preset target voice pattern includes: Real-time collecting the voice data of the driver through the voice collecting device of the first vehicle, and performing voice analysis on the collected voice data to extract the voice characteristics; Using a preset voice database to match the voice characteristics with the voice database, and determining whether the voice characteristics match the voice database; If not, controlling the voice collecting device to continue collecting the voice data of the driver.

5. The driving assistance method according to claim 2, wherein The vehicle environment information includes target detection data and image data; The vehicle-mounted sensor includes a vehicle-mounted radar sensor and a vehicle-mounted camera, and obtaining the vehicle environment information of the vehicle-mounted sensor includes: Obtaining the target detection data through the vehicle-mounted radar sensor, and obtaining the image data through the vehicle-mounted camera; Calculating a first state parameter based on the target detection data, and calculating a second state parameter through the image data; The first state parameter includes a first relative speed v1, a first acceleration a1, and a first relative distance d1, and the second state parameter includes a second relative speed v2, a second acceleration a2, and a second relative distance d2; Fusing the first state parameter and the second state parameter by using a weighted average or Kalman filtering algorithm to obtain the driving state parameter.

6. The driving assistance method according to claim 5, wherein Calculating the second state parameter through the image data includes: Extract the pixel height h of the target object in the image data p and the actual height h of the target object r , and obtain a second relative distance d2 through the pixel height h p and the actual height h r . The second relative distance d2 satisfies the following calculation formula: ； where f is a calibration parameter.

7. The driving assistance method according to claim 5, wherein: judging whether the first vehicle is in a dangerous driving environment according to the driving state parameters includes: calculating a danger state parameter S through the relative speed v and the relative distance d, and the danger state parameter is used to represent the driving state of the first vehicle: wherein, when the value of the danger state parameter S is 1, it is determined that the first vehicle is in a dangerous driving environment; when the value of the danger state parameter S is 0, it is determined that the first vehicle is not in a dangerous driving environment.

8. A driving assistance system for a vehicle, characterized in that, The driving assistance method according to any one of claims 1-7 is applied to the driving assistance system, and the driving assistance system includes: a voice module, which is used to obtain the voice characteristics of the driver and judge whether the voice characteristics fall into a preset voice database; a judgment module, which is used to judge whether the first vehicle is in a dangerous driving environment: an execution module, which is used to calculate the braking timing and braking force of the first vehicle and control the first vehicle to brake.

9. A vehicle, characterized in that, The driving assistance method according to any one of claims 1-8 is applied to the vehicle.

Images