Vehicle decision intention recognition method considering driving style
By introducing driving style classification and game theory methods, constructing a vehicle decision intention combination, and dynamically adjusting the vehicle decision weight, the problem of existing technologies ignoring personalized driving styles is solved, achieving more accurate and faster vehicle decisions, and improving the safety and comfort of intelligent driving.
Patent Information
- Application Number
- CN202511049455.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-29
- Publication Date
- 2025-09-09
AI Technical Summary
Existing vehicle decision-making systems rely on environmental sensor data and fixed rules, ignoring the driver's personalized driving style, resulting in reduced decision-making accuracy and stability in complex and dynamic driving scenarios. In particular, they lack consideration of individual differences when responding to emergencies, affecting driving safety and comfort.
By introducing driving style classifications (aggressive, smooth, cautious) and their corresponding personality factors, nine decision intention combinations of target vehicles and interacting vehicles are constructed. Game theory methods are used to solve the optimal decision. Combined with dynamic weights and time difference mechanisms, the predicted time and risk score of vehicles arriving at intersections are updated in real time, and the vehicle decision intention assessment weights are dynamically adjusted.
It significantly improves the recognition accuracy of vehicle decision-making intentions, reduces the misjudgment rate of intentions, enhances the rationality and rapid response capabilities of decisions, adapts to complex and changing traffic environments, and improves the safety and user experience of intelligent driving.
Smart Images
Figure CN120606844A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent vehicle technology, and in particular to a method for identifying vehicle decision intention based on driving style. Background Art
[0002] With the rapid development of autonomous driving and intelligent transportation technologies, the intelligence level of vehicles continues to increase. However, existing vehicle decision-making systems mostly rely on environmental sensor data and preset driving rules, requiring large amounts of data processing. This requires high computing power and is not easily deployable. Traditional driving intention recognition methods are typically based on fixed rules and models, focusing on acquiring and analyzing information about the vehicle's surroundings while ignoring the driver's individual driving style. This approach can lead to reduced decision-making accuracy and stability when handling complex and dynamic driving scenarios. In particular, when responding to emergencies, complex road environments, and changing traffic conditions, the lack of consideration for individual driver differences often results in inaccurate or inefficient vehicle behavior, impacting driving safety and comfort. Furthermore, while existing research has explored the identification and classification of driver behavior patterns, there are still deficiencies in effectively applying these patterns to intelligent vehicle decision-making. Therefore, developing innovative methods that can incorporate individual driving styles and improve decision-making intention recognition and adaptability is crucial for enhancing intelligent driving safety and user experience. Summary of the Invention
[0003] In order to address the shortcomings of the above-mentioned existing technologies, the present invention proposes a vehicle decision intention recognition method that takes driving style into consideration, in order to integrate the driver's individual driving style into the vehicle's intention decision-making, thereby improving the intelligent vehicle's ability to understand and predict the driver's intention, and optimizing the performance of the vehicle's decision support and assisted driving systems.
[0004] In order to achieve the above-mentioned object, the present invention adopts the following technical solutions: The method for identifying vehicle decision intention considering driving style of the present invention is characterized in that it comprises the following steps: Step 1: Represents the interval between adjacent moments, and the total observation time is , let the current time be ,and ; Get Vehicle trajectory information at each moment, including: Time target vehicle Location ,speed , and the target vehicle Interacting vehicles with intersecting trajectories Location ,speed , the intersection of the two vehicle trajectories ; Step 2: Make the target vehicle Interactive vehicles There are three types of driving decision intentions, including acceleration decision intention, uniform speed decision intention, and deceleration decision intention. Let the vehicle acceleration corresponding to acceleration decision intention, uniform speed decision intention, and deceleration decision intention be 、 、 ; The target vehicle The acceleration decision intention, constant speed decision intention and deceleration decision intention in the driving decision intention are respectively expressed as 、 、 Indicates that interactive vehicles The acceleration decision intention, constant speed decision intention and deceleration decision intention in the driving strategy intention are respectively expressed as 、 、 Indicates that the target vehicle with 9 combinations is constructed and interactive vehicles Driving decision intention group ;in, represents any driving decision intention group; Step 3: Calculate the target vehicle Interactive vehicles exist Safety score and speed score of the driving decision intention group at the moment; Step 4. Set the driving style category, including: aggressive, smooth and cautious; Calculate target vehicles based on driving style categories Interactive vehicles The relative strengths of driving styles , and interactive vehicles With the target vehicle The relative strengths of driving styles ; Step 5: Calculate the target vehicle based on the safety score and speed score exist Always make the first decision intention The comprehensive score of and interactive vehicles exist Always make the first Driving decision intention The comprehensive score of ; Step 6: Determine the target vehicle by formula (9) and interactive vehicles The optimal driving decision intention group ; (9) In formula (9), argmax is the maximum value function, Indicates the target vehicle exist The best decision at any time, Represents interactive vehicles exist The best decision at the moment; Step 7: Assign to After that, return to step 3 and execute sequentially until t>T. Finally, the vehicle's driving decision intention strategy within a period of time is obtained.
[0005] The method for identifying vehicle decision intention considering driving style according to the present invention is also characterized in that step 3 includes the following steps: Step 3.1: Calculate the target vehicle according to formula (1) exist Always make the first Driving decision intention Estimated arrival time at the intersection of the trajectories Time , interactive vehicles exist Always make the first Driving decision intention Estimated arrival time at the intersection of the trajectories Time ; (1) In formula (1), , Target vehicles , interactive vehicles exist The speed of time, , Target vehicles , interactive vehicles exist Time to the intersection of the two vehicle trajectories The distance value, Target vehicle exist Always make the first decision intention The corresponding acceleration value is For interactive vehicles exist Always make the first Driving decision intention The corresponding acceleration value when ; Step 3.2: Calculate the target vehicle according to formula (2) exist Always make the first decision intention Estimated arrival time at the intersection of the trajectories Time difference , interactive vehicles exist Always make the first Driving decision intention Estimated arrival time at the intersection of the trajectories Time difference ; (2) Step 3.3: Calculate the target vehicle according to formula (3) exist Always make the first decision intention Safety score at , and interactive vehicles exist Always make the first Driving decision intention Safety score at ; (3) In formula (3), , Target vehicles , interactive vehicles exist Estimated arrival trajectory intersection when no decision is made The time difference, is the corresponding weight coefficient; Step 3.4: Calculate the target vehicle according to formula (4) exist Always make the first decision intention hour The speed of time , and interactive vehicles exist Always make the first decision intention hour The speed of time ; (4) Step 3.5: Calculate the target vehicle according to formula (5) exist Always make the first decision intention Speed score , and interactive vehicles exist Always make the first Driving decision intention Speed score ; (5) In formula (5), is the road speed limit, Represent the corresponding weight coefficients respectively.
[0006] Furthermore, in step 4, the personality factor corresponding to any type of driving style is recorded as ; For the The personality factor corresponding to the driving style is calculated according to formula (6): as well as ; (6) In formula (6), Target vehicle No. Personality factors corresponding to driving styles, For interactive vehicles No. Personality factors corresponding to different driving styles.
[0007] Furthermore, in step 5, according to formula (7) as well as ; (7) In formula (7), , Target vehicle exist Always make the first decision intention Safety score at and target vehicle exist Always make the first decision intention Speed score The standardized results, , For interactive vehicles exist Always make the first Driving decision intention Safety score at and interactive vehicles exist Always make the first Driving decision intention Speed score The standardized results, Target vehicle exist Always make the first decision intention The equilibrium coefficient when For interactive vehicles exist Always make the first Driving decision intention The equilibrium coefficient is: (8).
[0008] The electronic device of the present invention includes a memory and a processor, and is characterized in that the memory is used to store a program that supports the processor to execute the vehicle decision intention recognition method, and the processor is configured to execute the program stored in the memory.
[0009] The present invention provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and the computer program executes the steps of the vehicle decision intention recognition method when the computer program is run by a processor.
[0010] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention dynamically adjusts the evaluation weight of vehicle decision-making intentions by incorporating driving style categories (aggressive, steady, cautious) and their corresponding personality factors. Compared to traditional methods that rely solely on fixed rules or environmental data, this invention's dynamic integration of driving styles more accurately reflects the individual behavior patterns of different drivers, significantly reducing the rate of misjudgment and improving the accuracy of intention recognition.
[0011] 2. This invention constructs nine combinations of decision intentions for the target vehicle and interacting vehicles, comprehensively calculates safety scores, speed scores, and relative driving styles, and employs game theory to determine the optimal decision. Compared to existing techniques that predict single vehicle intentions, this invention fully considers the impact of interacting vehicles' behavior and employs multi-dimensional game optimization, enhancing decision rationality and making decisions more consistent with the dynamic game characteristics of real-world traffic scenarios.
[0012] 3. This invention introduces a time difference and dynamic weighting mechanism to update the predicted time of arrival of the vehicle at the intersection and the risk score in real time. Combined with the dynamic adjustment of driving style, it can quickly respond to emergencies (such as emergency lane changes or sudden braking), thereby better adapting to complex and changing traffic environments. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1 It is the overall flow chart of the present invention; Figure 2 This is one of the example scene graphs of the embodiment of the present invention; Figure 3 Comparison diagram of the estimated speed and actual speed of the target vehicle and the interacting vehicle. DETAILED DESCRIPTION
[0014] In this embodiment, a vehicle decision intention recognition method that considers driving style is designed to incorporate the driver's individual driving style. By finding the optimal solution for the game between the vehicle and surrounding vehicles, the vehicle's driving intention recognition problem when interacting with other surrounding vehicles during driving is solved. The driver's driving style is incorporated into the multi-party game to improve the real-time recognition accuracy in complex situations, thereby enhancing the intelligent vehicle's ability to understand and predict the driver's intentions, thereby optimizing the performance of the vehicle's decision support and assisted driving system. Specifically, if Figure 1 As shown, the method includes: Step 1: Represents the interval between adjacent moments, and the total observation time is , let the current time be ,and ; Get Vehicle trajectory information at each moment, including: Time target vehicle Location ,speed , and the target vehicle Interacting vehicles with intersecting trajectories Location ,speed , the intersection of the two vehicle trajectories ; Step 2: Make the target vehicle Interactive vehicles There are three types of driving decision intentions, including acceleration decision intention, uniform speed decision intention, and deceleration decision intention. Let the vehicle acceleration corresponding to acceleration decision intention, uniform speed decision intention, and deceleration decision intention be 、 、 During the game, the accelerations of the aggressive, steady, and cautious drivers when they choose to accelerate through the expected conflict intersection are 、 、 ; Choose the uniform speed strategy ; The deceleration rates when deceleration is selected are 、 、 .
[0015] The target vehicle The acceleration decision intention, constant speed decision intention and deceleration decision intention in the driving decision intention are respectively expressed as 、 、 Indicates that interactive vehicles The acceleration decision intention, constant speed decision intention and deceleration decision intention in the driving strategy intention are respectively expressed as 、 、 Indicates that the target vehicle with 9 combinations is constructed and interactive vehicles Driving decision intention group ;in, Represents any driving decision intention group.
[0016] Step 3: Calculate the target vehicle Interactive vehicles exist The decision score of the driving decision intention group at the moment; Step 3.1: Calculate the target vehicle according to formula (1) exist Always make the first Driving decision intention Estimated arrival time at the intersection of the trajectories Time , interactive vehicles exist Always make the first Driving decision intention Estimated arrival time at the intersection of the trajectories Time ; (1) In formula (1), , Target vehicles , interactive vehicles exist The speed of time, , Target vehicles , interactive vehicles exist Time to the intersection of the two vehicle trajectories The distance value, Target vehicle exist Always make the first decision intention The corresponding acceleration value is For interactive vehicles exist Always make the first Driving decision intention The acceleration value corresponding to the time. In order to reduce the error, the speed here is calculated based on the actual speed.
[0017] Step 3.2: Calculate the target vehicle according to formula (2) exist Always make the first decision intention Estimated arrival time at the intersection of the trajectories Time difference , interactive vehicles exist Always make the first Driving decision intention Estimated arrival time at the intersection of the trajectories Time difference ; (2) Step 3.3: Calculate the target vehicle according to formula (3) exist Always make the first decision intention Safety score at , and interactive vehicles exist Always make the first Driving decision intention Safety score at ; (3) In formula (3), , Target vehicles , interactive vehicles exist Estimated arrival trajectory intersection when no decision is made The time difference, is the corresponding weight coefficient, which is 0.5 here.
[0018] Step 3.4: Calculate the target vehicle according to formula (4) exist Always make the first decision intention hour, The speed of time , and interactive vehicles exist Always make the first decision intention hour, The speed of time ; (4) Step 3.5: Calculate the target vehicle according to formula (5) exist Always make the first decision intention Speed score , and interactive vehicles exist Always make the first Driving decision intention Speed score ; (5) In formula (5), is the road speed limit, Represent the corresponding weight coefficients, here Take 0.5 for both. Take 0.45.
[0019] Step 4: Set the driving style categories, including aggressive, smooth and cautious; and record the personality factor corresponding to any driving style as ; For the Personality factors corresponding to driving styles, For cautious drivers, For stable drivers, For aggressive drivers; According to formula (6), the target vehicle Interactive vehicles The relative strengths of driving styles , and interactive vehicles With the target vehicle The relative strengths of driving styles ; (6) In formula (6), Target vehicle No. Personality factors corresponding to driving styles, For interactive vehicles No. Personality factors corresponding to different driving styles.
[0020] Step 5: Calculate the target vehicle according to formula (7) exist Always make the first decision intention The comprehensive score of , and interactive vehicles exist Always make the first Driving decision intention The comprehensive score of ; (7) In formula (7), , for and The standardized results, , for and The standardized results, Target vehicle exist Always make the first decision intention The equilibrium coefficient when For interactive vehicles exist Always make the first Driving decision intention The equilibrium coefficient is: (8) Step 6: Determine the target vehicle by formula (9) and interactive vehicles The optimal driving decision intention group ; (9) In formula (9), argmax is the maximum value function, Indicates the target vehicle exist The best decision at any time, Represents interactive vehicles exist The best decision at any time.
[0021] A trajectory intersection case is selected to illustrate the process of identifying vehicle decision intentions, such as Figure 2 As shown, the initial motion state of the two vehicles at the beginning of the first time step, for the target vehicle: , , , for interactive vehicles: , , According to the global steps described above, the total benefit function matrix of the two drivers at the first time step can be obtained, as shown in Table 1 below.
[0022] Table 1 The vehicle's profit function matrix at the first moment Step 7: After assigning the value to t, return to step 3 and execute sequentially until t>T. Finally, the driving decision intention strategy of the vehicle in a period of time is obtained, as shown in Table 2. The target vehicle Continuously accelerate while driving, interact with vehicles The process of slowing down.
[0023] Table 2 The best strategy combination of the two cars at all times Figure 3 To compare the simulated speed with the actual speed, the target vehicle is shown in the figure. During constant acceleration in driving, the interactive vehicle The results show that the estimated speeds of the two vehicles are almost consistent with their actual speeds.
[0024] In this embodiment, an electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.
[0025] In this embodiment, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are executed.
Claims
1. A vehicle decision intention recognition method considering driving style, characterized in that: The following steps are involved: Step 1: Represents the interval between adjacent moments, and the total observation time is , let the current time be ,and ; Get Vehicle trajectory information at each moment, including: Time target vehicle Location ,speed , and the target vehicle Interacting vehicles with intersecting trajectories Location ,speed , the intersection of the two vehicle trajectories ; Step 2: Make the target vehicle Interactive vehicles There are three types of driving decision intentions, including acceleration decision intention, uniform speed decision intention, and deceleration decision intention. Let the vehicle acceleration corresponding to acceleration decision intention, uniform speed decision intention, and deceleration decision intention be 、 、 ; The target vehicle The acceleration decision intention, constant speed decision intention and deceleration decision intention in the driving decision intention are respectively expressed as 、 、 Indicates that interactive vehicles The acceleration decision intention, constant speed decision intention and deceleration decision intention in the driving strategy intention are respectively expressed as 、 、 Indicates that the target vehicle with 9 combinations is constructed and interactive vehicles Driving decision intention group ;in, represents any driving decision intention group; Step 3: Calculate the target vehicle Interactive vehicles exist Safety score and speed score of the driving decision intention group at the moment; Step 4. Set the driving style category, including: aggressive, smooth and cautious; Calculate target vehicles based on driving style categories Interactive vehicles The relative strengths of driving styles , and interactive vehicles With the target vehicle The relative strengths of driving styles ; Step 5: Calculate the target vehicle based on the safety score and speed score exist Always make the first decision intention The comprehensive score of and interactive vehicles exist Always make the first Driving decision intention The comprehensive score of ; Step 6: Determine the target vehicle by formula (9) and interactive vehicles The optimal driving decision intention group ; (9) In formula (9), argmax is the maximum value function, Indicates the target vehicle exist The best decision at any time, Represents interactive vehicles exist The best decision at the moment; Step 7: Assign to After that, return to step 3 and execute sequentially until t>T. Finally, the vehicle's driving decision intention strategy within a period of time is obtained.
2. The method for identifying vehicle decision intention considering driving style according to claim 1, characterized in that: The step 3 comprises the following steps: Step 3.1: Calculate the target vehicle according to formula (1) exist Always make the first Driving decision intention Estimated arrival time at the intersection of the trajectories Time , interactive vehicles exist Always make the first Driving decision intention Estimated arrival time at the intersection of the trajectories Time ; (1) In formula (1), , Target vehicles , interactive vehicles exist The speed of time, , Target vehicles , interactive vehicles exist Time to the intersection of the two vehicle trajectories The distance value, Target vehicle exist Always make the first decision intention The corresponding acceleration value is For interactive vehicles exist Always make the first Driving decision intention The corresponding acceleration value when ; Step 3.2: Calculate the target vehicle according to formula (2) exist Always make the first decision intention Estimated arrival time at the intersection of the trajectories Time difference , interactive vehicles exist Always make the first Driving decision intention Estimated arrival time at the intersection of the trajectories Time difference ; (2) Step 3.3: Calculate the target vehicle according to formula (3) exist Always make the first decision intention Safety score at , and interactive vehicles exist Always make the first Driving decision intention Safety score at ; (3) In formula (3), , Target vehicles , interactive vehicles exist Estimated arrival trajectory intersection when no decision is made The time difference, is the corresponding weight coefficient; Step 3.4: Calculate the target vehicle according to formula (4) exist Always make the first decision intention hour The speed of time , and interactive vehicles exist Always make the first decision intention hour The speed of time ; (4) Step 3.5: Calculate the target vehicle according to formula (5) exist Always make the first decision intention Speed score , and interactive vehicles exist Always make the first Driving decision intention Speed score ; (5) In formula (5), is the road speed limit, Represent the corresponding weight coefficients respectively.
3. The method for identifying vehicle decision intention considering driving style according to claim 2, characterized in that: In step 4, the personality factor corresponding to any type of driving style is recorded as ; For the The personality factor corresponding to the driving style is calculated according to formula (6): as well as ; (6) In formula (6), Target vehicle No. Personality factors corresponding to driving styles, For interactive vehicles No. Personality factors corresponding to different driving styles.
4. The method for identifying vehicle decision intention considering driving style according to claim 3, characterized in that: In step 5, according to formula (7), as well as ; (7) In formula (7), , Target vehicle exist Always make the first decision intention Safety score at and target vehicle exist Always make the first decision intention Speed score The standardized results, , For interactive vehicles exist Always make the first Driving decision intention Safety score at and interactive vehicles exist Always make the first Driving decision intention Speed score The standardized results, Target vehicle exist Always make the first decision intention The equilibrium coefficient when For interactive vehicles exist Always make the first Driving decision intention The equilibrium coefficient is: (8)。 5. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store a program that supports the processor to execute the vehicle decision intention recognition method according to any one of claims 1 to 4, and the processor is configured to execute the program stored in the memory.
6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the vehicle decision intention recognition method according to any one of claims 1 to 4 are executed.