Vehicle control method and device based on man-machine trust, medium and program product

Abstract

The invention discloses a vehicle control method and device based on man-machine trust, a medium and a program product, and relates to the technical field of intelligent driving, and the method comprises the steps: determining a man-machine trust value according to the obtained real-time overtaking data of a current vehicle; determining a human control weight of the current vehicle according to the man-machine trust value; and based on the obtained human control instruction, the machine control instruction and the human control weight, the current vehicle is controlled to run. According to the method, the human control weight of the automatic driving system of the current vehicle is dynamically adjusted according to the real-time overtaking performance, the defect that a driver suddenly takes over the vehicle with full right is avoided, and therefore the risk of traffic accidents when the vehicle is controlled through a man-machine hybrid system is reduced.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, and particularly to a vehicle control method, device, medium, and program product based on human-machine trust. Background Art

[0002] With the development of computer and artificial intelligence technologies, driving assistance systems have become an effective way to improve traffic safety and reduce the workload of drivers. However, due to complex traffic environments and legal issues, fully autonomous driving is difficult to achieve in the short term, and the automation level of intelligent vehicles remains in the human-machine hybrid stage before full automation.

[0003] Most current human-machine hybrid systems use intervention control methods. When sensors or decision-making algorithms installed in the vehicle detect potential dangers, the driving right is directly transferred to the driver. However, when the driver has not been in the driving loop for a long time, they may be unable to quickly and effectively take over the vehicle due to the lack of perception of the current traffic situation and vehicle operation mode, thus increasing the probability of traffic accidents. Summary of the Invention

[0004] The main purpose of this application is to provide a vehicle control method, device, medium, and program product based on human-machine trust, aiming to reduce the risk of traffic accidents when controlling vehicles through human-machine hybrid systems.

[0005] To achieve the above purpose, this application proposes a vehicle control method based on human-machine trust, and the method includes:

[0006] Determine the human-machine trust value according to the obtained real-time overtaking data of the current vehicle;

[0007] Determine the human control weight of the current vehicle according to the human-machine trust value;

[0008] Control the current vehicle to travel based on the obtained human control instructions, machine control instructions, and the human control weight.

[0009] In one embodiment, the step of determining the human-machine trust value according to the obtained real-time overtaking data of the current vehicle includes:

[0010] Determine the human subjective trust value according to the real-time overtaking data;

[0011] Determine the human-machine trust value according to the human subjective trust value and the preset machine objective ability value.

[0012] In one embodiment, the step of determining the human subjective trust value according to the real-time overtaking data includes:

[0013] Determine the real-time performance score of the current vehicle according to the real-time overtaking data;

[0014] Determine the human subjective trust value according to the historical subjective trust value and the real-time performance score.

[0015] In one embodiment, the step of determining the human subjective trust value according to the historical subjective trust value and the real-time performance score includes:

[0016] Obtain offline operation data, where the offline operation data includes the current trust value, the historical trust value, and the offline overtaking performance score;

[0017] By the least squares method, fit the current trust value, the historical trust value, and the offline overtaking performance score to determine the calculation weight coefficient of the human subjective trust value;

[0018] Determine the human subjective trust value according to the historical subjective trust value, the real-time performance score, and the calculation weight coefficient.

[0019] In one embodiment, the method further includes:

[0020] In the case of determining the human subjective trust value for the first time, determine the historical subjective trust value according to the obtained machine uncertainty, machine transparency, and human conservatism value.

[0021] In one embodiment, the real-time overtaking data includes: initial distance, parallel boundary distance, lateral acceleration, overtaking distance, minimum distance, initial speed, and initial acceleration. The step of determining the real-time performance score according to the real-time overtaking data includes:

[0022] Determine the first index score of the initial distance according to the initial distance, the initial speed, the initial acceleration, a preset safety distance, and the reaction time;

[0023] Determine the second index score of the parallel boundary distance according to the width of the one-way lane, the width of the current vehicle, and the parallel boundary distance;

[0024] Determine the third index score of the overtaking distance according to the overtaking distance and the safety distance;

[0025] Respectively determine the fourth index score of the minimum distance and the fifth index score of the lateral acceleration;

[0026] Determine the real-time performance score according to the first index score, the second index score, the third index score, the fourth index score, and the fifth index score.

[0027] In one embodiment, the method further includes:

[0028] In the case where the human - machine trust value is less than or equal to the first preset threshold, keep the human control weight as the preset first weight value;

[0029] In the case where the human - machine trust value is greater than the first preset threshold and less than the second preset threshold, determine the human control weight as the second weight value, where the second weight value has an inverse proportional relationship with the human - machine trust value;

[0030] In the case where the human - machine trust value is greater than or equal to the second preset threshold, keep the human control weight as the preset third weight value, where the first weight value is greater than the third weight value.

[0031] In addition, to achieve the above - mentioned purpose, the present application also proposes a vehicle control device based on human - machine trust. The vehicle control device based on human - machine trust includes:

[0032] A trust value determination module, configured to determine the human - machine trust value according to the acquired real - time overtaking data of the current vehicle;

[0033] A weight determination module, configured to determine the human control weight of the current vehicle according to the human - machine trust value;

[0034] A vehicle control module, configured to control the current vehicle to travel based on the acquired human control instruction, machine control instruction, and the human control weight.

[0035] In addition, to achieve the above - mentioned purpose, the present application also proposes an electronic device. The device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor. The computer program is configured to implement the steps of the vehicle control method based on human - machine trust as described above.

[0036] In addition, to achieve the above - mentioned purpose, the present application also proposes a storage medium. The storage medium is a computer - readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the vehicle control method based on human - machine trust as described above.

[0037] In addition, to achieve the above - mentioned purpose, the present application also provides a computer program product. The computer program product includes a computer program. When the computer program is executed by a processor, it implements the steps of the vehicle control method based on human - machine trust as described above.

[0038] One or more technical solutions proposed in this application have at least the following technical effects: First, according to the obtained real-time overtaking data of the current vehicle, the human-machine trust value is determined to evaluate the trust state of humans in autonomous driving control in the current overtaking situation. Furthermore, based on the real-time determined human-machine trust value, the human control weight of the current vehicle is determined. During the driving process of the vehicle, the control authority of the human driver over the vehicle is reasonably adjusted according to the degree of trust. For example, when the trust value is high, the human control weight is appropriately reduced, and vice versa, to achieve dynamic allocation of authority. This flexible control architecture effectively avoids the defect of direct full transfer of traditional driving assistance systems and realizes a smooth transition of human-machine decision-making through weight adjustment. Finally, based on the obtained human control instructions, machine control instructions, and the determined human control weight, the vehicle is controlled to travel, effectively solving the traffic safety problems that may be caused by unreasonable authority allocation and poor cooperation between the driver and the system in the human-machine hybrid control system, thereby reducing the risk of traffic accidents when controlling the vehicle through the human-machine hybrid system and improving the safety and reliability of intelligent driving in the overtaking and lane-changing scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The accompanying drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with this application and, together with the specification, are used to explain the principles of this application.

[0040] To more clearly illustrate the technical solutions in the embodiments of this application or in the prior art, the following will briefly introduce the accompanying drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0041] Figure 1 It is a schematic flowchart provided for Embodiment 1 of the vehicle control method based on human-machine trust in this application;

[0042] Figure 2 It is a schematic overall flowchart provided for Embodiment 2 of this application to determine the human subjective trust value;

[0043] Figure 3 It is a schematic overall framework diagram of the vehicle control method based on human-machine trust provided for Embodiment 2 of this application;

[0044] Figure 4 It is a schematic module structure diagram of the vehicle control device based on human-machine trust for the embodiments of this application;

[0045] Figure 5 It is a schematic device structure diagram of the hardware operating environment involved in the vehicle control method based on human-machine trust in the embodiments of this application.

[0046] The realization of the purpose, functional features and advantages of this application will be further described in conjunction with embodiments with reference to the accompanying drawings. Specific Embodiments

[0047] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of this application and are not used to limit this application.

[0048] To better understand the technical solutions of this application, the following will be described in detail in conjunction with the accompanying drawings of the specification and specific embodiments.

[0049] In this embodiment, for the convenience of description, the vehicle terminal is used as the execution subject for elaboration below.

[0050] Since human trust is difficult to measure, and it is difficult to establish a model suitable for controller design that simultaneously satisfies accuracy, interpretability, and predictability, there is almost no human-machine trust model for overtaking and lane-changing scenarios currently; and the human-machine control method through intervention control has the risk of increasing the probability of traffic accidents due to the driver's lack of situation awareness.

[0051] This application provides a solution. First, according to the obtained real-time overtaking data of the current vehicle, determine the human-machine trust value to evaluate the trust state of humans in autonomous driving control in the current overtaking situation; furthermore, based on the real-time determined human-machine trust value, determine the human control weight of the current vehicle. During the driving process of the vehicle, reasonably adjust the control authority of the human driver over the vehicle according to the degree of trust. For example, when the trust value is high, appropriately reduce the human control weight, and vice versa, to achieve dynamic allocation of authority. This elastic control architecture effectively avoids the defect of the direct full transfer of traditional driving assistance systems and realizes a smooth transition of human-machine decision-making through weight adjustment; finally, based on the obtained human control instructions, machine control instructions, and determined human control weight, control the vehicle to drive, effectively solving the traffic safety problems that may be caused by unreasonable authority allocation and poor cooperation between the driver and the system in the human-machine hybrid control system, thereby reducing the risk of traffic accidents when controlling the vehicle through the human-machine hybrid system and improving the safety and reliability of intelligent driving in the overtaking and lane-changing scenarios.

[0052] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication, and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions. The following takes the vehicle terminal as the execution subject as an example to illustrate this embodiment and the following embodiments.

[0053] Based on this, the embodiments of this application provide a vehicle control method based on human-machine trust, referring to Figure 1 , Figure 1This is a schematic flowchart of the first embodiment of the vehicle control method based on human-machine trust in this application.

[0054] In this embodiment, the vehicle control method based on human-machine trust includes steps S10 to S30:

[0055] Step S10, determine the human-machine trust value according to the obtained real-time overtaking data of the current vehicle;

[0056] In a feasible embodiment, by quantifying the overtaking performance of the current vehicle, a trust evaluation model (human-machine trust expression) for human-machine collaborative driving is established to provide a basis for subsequent control authority allocation; furthermore, based on the real-time overtaking data, the human-machine trust value under the current overtaking performance can be determined, so as to achieve more reasonable and safer vehicle control.

[0057] Optionally, the real-time overtaking data refers to a series of data sets continuously collected by vehicle sensors during the overtaking process, including vehicle speed, distance from the vehicle ahead, lane departure situation, etc., providing instant environmental perception information for the human-machine hybrid automatic driving system. And the human-machine trust value is used to measure the degree of trust of the human driver in the automatic driving system under the current overtaking performance.

[0058] Optionally, the human-machine trust value can also be predicted through a specific algorithm model. For example, using a deep learning model, features of the real-time overtaking data, such as acceleration features and trajectory change features, are extracted, and based on the trust value judgment criteria learned during the training process, a human-machine trust value reflecting the current overtaking performance is generated.

[0059] In a feasible implementation manner, step S10 includes:

[0060] Step S11, determine the human subjective trust value according to the real-time overtaking data;

[0061] Optionally, the human subjective trust value reflects the degree of recognition of the driver for the overtaking performance of the current vehicle, and it can also be determined by constructing an expression or through an algorithm model.

[0062] Optionally, the subjective trust value of the driver for the automatic driving system can be determined by combining the real-time overtaking data and the driver operation feedback; for example, the minimum distance from the vehicle ahead can be extracted from the real-time overtaking data, and the number and depth of brake pedal presses can be extracted from the driver operation feedback, and based on this, the overtaking performance of the current automatic driving system is scored to determine the human subjective trust value.

[0063] Step S12, determine the human-machine trust value according to the human subjective trust value and the preset machine objective ability value.

[0064] Optionally, the machine objective ability value refers to the objective performance index generated by the vehicle-mounted system through a self-check program, which will not fluctuate significantly in a short period of time and can be updated within a preset time range.

[0065] Optionally, the human-machine trust value can be determined by weighted fusion according to the weight distribution between the human subjective trust value and the machine objective ability value. For example, according to the current road conditions (such as the number of vehicles, the number of lanes, etc.), the weight coefficient between the two can be determined. In the case of a large number of vehicles and a small number of lanes, the weight coefficient of the machine objective ability value can be appropriately reduced, and more reliance on the human subjective trust value is used to determine the human-machine trust value that affects the final weight distribution.

[0066] Optionally, the human-machine trust value is determined according to the ratio between the human subjective trust value and the machine objective ability value.

[0067] Exemplarily, the human-machine trust value has the following expression:

[0068]

[0069] where k represents the time window, the size of which is not necessarily uniform, but is determined according to the real-time overtaking time of the vehicle; and represent the machine objective ability value and the human subjective trust value respectively. When exceeds 1, it can be determined that the driver is in a state of over-trust at this time, and the driver is reminded. The driver can reduce the human subjective trust value and thus the human-machine trust value by adjusting the vehicle speed, driving direction, etc.

[0070] In this embodiment, by comprehensively considering the human subjective trust and the objective ability of the vehicle to determine the human-machine trust value, ensuring that the vehicle-mounted system reasonably distributes permissions within its own performance range through the objective ability of the machine, and at the same time fully considering the subjective score of the current overtaking performance, it helps to achieve more accurate human-machine collaborative control and reduce the risk of traffic accidents caused by inaccurate human-machine trust assessment.

[0071] Step S20, determine the human control weight of the current vehicle according to the human-machine trust value;

[0072] Optionally, the human control weight refers to the proportion of the driver's operation in vehicle driving control.

[0073] Optionally, the human control weight of the current vehicle can be determined according to the mapping relationship between the human-machine trust value and the preset trust value and control weight. For example, when the human-machine trust value is high, the mapped human control weight is low, allowing the autonomous driving system to dominate the vehicle driving; while when the human-machine trust value is low, the mapped human control weight will increase accordingly, enabling the driver to participate more in vehicle control to handle possible complex or emergency road conditions.

[0074] In a feasible implementation manner, the vehicle control method based on human-machine trust further includes:

[0075] Step S21, in the case where the human-machine trust value is less than or equal to the first preset threshold, keep the human control weight as the preset first weight value;

[0076] Step S22, in the case where the human-machine trust value is greater than the first preset threshold and less than the second preset threshold, determine the human control weight as the second weight value, where the second weight value is inversely proportional to the human-machine trust value;

[0077] Step S23, in the case where the human-machine trust value is greater than or equal to the second preset threshold, keep the human control weight as the preset third weight value, where the first weight value is greater than the third weight value.

[0078] Optionally, the first preset threshold is used to define the lower limit of the human-machine trust value, marking the lower boundary of the human-machine cooperation trust level; the second preset threshold is used to define the upper limit of the human-machine trust value, marking the higher boundary of the human-machine cooperation trust level.

[0079] In a feasible embodiment, in the case where the human-machine trust value is less than or equal to the first preset threshold, that is, when the human does not trust the autonomous driving, although the human is given greater control authority, the machine also retains a certain degree of control, which can relieve the driving pressure for users to a certain extent; while in the case where the human-machine trust value is greater than or equal to the second preset threshold, that is, when the driver highly trusts the autonomous driving system, the system still receives the driver's control input instead of being directly fully controlled by the system, so that the driver always has partial situation awareness ability, thereby reducing the risk of traffic accidents when controlling the vehicle through the human-machine hybrid system; and in the case where the human-machine trust value is greater than the first preset threshold and less than the second preset threshold, the higher the human trust in the autonomous driving, that is, the greater the human-machine trust value, the lower the human control weight, and the vehicle is more automatically controlled by the machine.

[0080] Exemplarily, according to the human-machine trust value to determine the allocation function of the human control weight λ:

[0081]

[0082] Among them, 0.8 is the preset first weight value, and 0.2 is the preset third weight value; T1 and T2 respectively represent the first preset threshold and the second preset threshold. Among them, T1 is a relatively small number (usually less than 0.5), indicating that humans do not trust the autonomous driving system. When the human-machine trust value is less than this value, humans need to be given greater vehicle control authority and reminded that they are in a state of lack of trust, and they can appropriately reduce the control of the vehicle; T2 is a relatively large number (usually greater than 1.2), indicating that humans overly trust the autonomous driving system. When the human-machine trust value exceeds this value, humans still need to be given a relatively small control authority and reminded that they are in a state of over-trust, and the supervision and control of the vehicle need to be strengthened. When the human-machine trust value is between T1 and T2, an inverse proportional linear function related to the human-machine trust value is used to determine the human control weight of the autonomous driving system.

[0083] Optionally, in addition to the human-machine trust value, other factors can also be combined, such as vehicle driving speed, road type, weather conditions, etc., to comprehensively adjust the human control weight. For example, under bad weather conditions, even if the human-machine trust value is high, the human control weight can be maintained at a relatively high value to ensure that the driver can promptly respond to situations such as slippery road surfaces caused by the weather, further reducing the risk of traffic accidents.

[0084] In this embodiment, the human control authority is dynamically adjusted based on the human-machine trust value. While reminding of the bad trust state, the conflict of the transfer of control rights between humans and machines is reduced, thereby reducing the risk of traffic accidents when controlling the vehicle through the human-machine hybrid system.

[0085] Step S30, control the current vehicle to drive based on the obtained human control instruction, machine control instruction, and human control weight.

[0086] In a feasible embodiment, the in-vehicle terminal determines specific action instructions through algorithm processing by comprehensively considering the human control instruction and the machine control instruction according to the human control weight, and realizes the adjustment of driving states such as the driving direction and speed of the vehicle.

[0087] Optionally, the human control instruction refers to the operation instruction input by the driver through sensors such as the steering wheel, accelerator, and brake, which reflects the driver's operation intention; while the machine control instruction refers to the driving decision automatically generated by the autonomous driving system for the current road conditions.

[0088] Exemplarily, the actual control input u of the vehicle is as follows:

[0089] u = λu h +(1 - λ)u m

[0090] Among them, λ is the human control weight, uh is a human control instruction, u m is a machine control instruction.

[0091] This embodiment provides a vehicle control method based on human-machine trust. By evaluating the degree of trust of humans in autonomous driving control in the current overtaking scenario and dynamically adjusting the proportion of the driver's control input in vehicle driving control according to this degree of trust, it effectively avoids the problem of suddenly transferring full control of the vehicle in traditional intervention control methods and reduces the risk of traffic accidents caused by poor human-machine cooperation.

[0092] Based on the first embodiment of this application, in the second embodiment of this application, the same or similar content as in the above-mentioned Embodiment 1 can be referred to the above introduction and will not be repeated hereinafter. On this basis, step S11 includes:

[0093] Step A10, determine the real-time performance score of the current vehicle according to the real-time overtaking data;

[0094] Optionally, the real-time performance score is used to reflect the comprehensive performance of the vehicle in aspects such as safety and comfort in the lane-changing overtaking scenario.

[0095] In a feasible implementation manner, the real-time overtaking data includes: initial distance, parallel boundary distance, lateral acceleration, overtaking distance, minimum distance, initial speed, and initial acceleration. Step A10 includes:

[0096] Step A11, determine the first index score of the initial distance according to the initial distance, initial speed, initial acceleration, preset safety distance, and reaction time;

[0097] Optionally, determine data such as the initial distance, initial speed, and initial acceleration from the real-time overtaking data, combine data such as the preset safety distance and reaction time, and substitute them into the preset scoring model for calculation to obtain the first index score. This scoring model can be a mathematical formula or a machine learning model, etc.

[0098] Optionally, the first index score can be determined according to the difference between the initial distance and its ideal value, where the ideal value of the initial distance is determined according to the initial speed, initial acceleration, preset safety distance, and reaction time.

[0099] Exemplarily, the mathematical formula of the first index score score1 is as follows:

[0100]

[0101] where, x initDenote the initial distance as \(d_0\), \(\mu_1\) as the ideal value of the initial distance, \(\sigma\) as an adjustable parameter used to measure the degree of dispersion of the distance. Through the processing of the exponential function and \(\sigma\), the deviation between the initial distance and the ideal distance can be converted into a value between 0 and 1, realizing the normalization of the index score; \(\nu\) is the initial speed, \(a\) is the initial acceleration, \(\Delta t\) is the preset reaction time of the driver, and \(d\) safe is the preset safety distance, and \(v\Delta t\) represents the driving distance of the vehicle from the moment the driver decides to start overtaking to the actual start of operation; That is, in the case of the vehicle in front suddenly braking, etc., it refers to the safety distance that needs to be reserved for the current vehicle to also stop immediately, so as to reduce the risk of traffic accidents such as rear-end collisions.

[0102] Step A12: Determine the second index score of the parallel boundary distance according to the width of the one-way lane, the width of the current vehicle, and the parallel boundary distance;

[0103] Optionally, the parallel boundary distance refers to the distance between the current vehicle and the road boundary line away from the vehicle being overtaken when the current vehicle and the vehicle being overtaken are driving in parallel during the overtaking process, and is used to measure whether the current vehicle has exited the lane, that is, to measure the safety of the current overtaking process.

[0104] Optionally, the second index score can be determined according to the difference between the parallel boundary distance and its ideal value, where the ideal value of the parallel boundary distance is determined according to the width of the one-way lane and the width of the current vehicle.

[0105] Exemplarily, the mathematical calculation formula of the second index score \(score2\) is as follows:

[0106]

[0107] where \(y\) boundary represents the parallel boundary distance, \(\mu_2\) represents the ideal value of the parallel boundary distance, \(\sigma\) is an adjustable parameter, and this second index score also realizes normalization through the exponential function; \(d\) road is the width of the one-way lane, \(d\) car is the width of the current vehicle.

[0108] Step A13: Determine the third index score of the overtaking distance according to the overtaking distance and the safety distance;

[0109] Optionally, the overtaking distance refers to the distance between the current vehicle and the vehicle being overtaken when the current vehicle returns to its original lane after completing overtaking.

[0110] Optionally, the third index score can be determined according to the difference between the overtaking distance and the preset safety distance.

[0111] Exemplarily, the mathematical calculation formula of the third index score \(score3\) is as follows:

[0112]

[0113] where x final represents the overtaking distance, and μ3 represents the ideal value of the overtaking distance; σ1 is for x final less than d safe a designed value. When the overtaking distance is less than the ideal value, too small an overtaking distance may increase the collision risk. Therefore, a relatively small σ1 can be used to more strictly penalize this situation, causing the corresponding third index score to decrease faster, emphasizing the importance of maintaining a sufficient safety distance in the overtaking performance score; σ2 is for x final greater than d safe a designed value. Mainly because when the overtaking distance is greater than the ideal value, although the distance is relatively farther and safer, it may affect other factors such as traffic efficiency. Therefore, a relatively large σ2 can be used to handle this situation relatively mildly, allowing a relatively large distance deviation within a certain range and still giving a relatively high score for a certain deviation in the overtaking distance, in order to reflect the balance between safety and efficiency in the overall overtaking process.

[0114] Step A14, respectively determine the fourth index score of the minimum distance and the fifth index score of the lateral acceleration;

[0115] Optionally, the minimum distance refers to the minimum distance between the current vehicle and the vehicle being overtaken during the entire overtaking process; the lateral acceleration refers to the maximum value of the lateral acceleration during the entire overtaking process.

[0116] Optionally, the fourth index score can be determined according to the minimum distance, where the fourth index score is in a direct proportional relationship with the minimum distance; the fifth index score can be determined according to the lateral acceleration, where the fifth index score is in an inverse proportional relationship with the lateral acceleration.

[0117] Exemplarily, the calculation formulas for the fourth index score score4 and the fifth index score score5 are as follows:

[0118]

[0119] where d min represents the minimum distance, a ymax represents the lateral acceleration, and the ideal values of both are 0; for the index of the minimum distance, the closer its value is to 0, the more dangerous the overtaking process of the current vehicle is, and the lower the corresponding fourth index score is; for the index of the lateral acceleration, the closer its value is to 0, the smoother the overtaking process of the current vehicle is, and the higher the corresponding fifth index score is.

[0120] Step A15: Determine the real-time performance score based on the first index score, the second index score, the third index score, the fourth index score, and the fifth index score.

[0121] Optionally, the real-time performance score can be obtained by calculating the weighted sum of the above five index scores; the weight indices of different indices can be determined according to the influence degree of each index score on the safety and comfort of the overtaking process, or can be allocated based on training the optimal weight combination in different traffic scenarios by a machine learning model. Therefore, the final real-time performance score Among them, w1, w2, w3, w4, and w5 respectively represent the weight indices corresponding to each index score.

[0122] In this embodiment, by comprehensively evaluating the scores of the five key indices including the initial distance, the parallel boundary distance, the overtaking distance, the minimum distance, and the lateral acceleration during the overtaking process, the real-time performance score of the vehicle can comprehensively reflect the safety, efficiency, and comfort during the overtaking process.

[0123] Step A20: Determine the human subjective trust value based on the historical subjective trust value and the real-time performance score.

[0124] Optionally, the historical subjective trust value reflects the driver's trust level in the vehicle's automatic driving system in the past, while the real-time performance score reflects the actual performance of the vehicle during the current overtaking process. By comprehensively considering these two factors to determine the human subjective trust value, it can avoid being overly trusting or completely lacking trust in the entire automatic driving system due to the current accidental performance.

[0125] Optionally, the historical subjective trust value and the implementation performance score can be comprehensively considered by means of weighted sum, non-linear function, machine learning model, etc. This embodiment does not make specific limitations on this.

[0126] In a feasible embodiment, the vehicle control method based on human-machine trust further includes:

[0127] Step A201: In the case of determining the human subjective trust value for the first time, determine the historical subjective trust value according to the obtained machine uncertainty, machine transparency, and human conservatism value.

[0128] Optionally, the machine uncertainty reflects the degree of confidence of the current vehicle's automatic driving system itself in environmental perception and decision-making. The machine transparency refers to the clarity of the driving information conveyed by the automatic driving system to the driver. These two are determined based on the current environmental information and the software and hardware information of the system itself; while the human conservatism value reflects the driving caution degree of the current driver, which can be manually input by the driver or determined according to the driver's historical driving data.

[0129] Optionally, the initially determined historical subjective trust value, i.e., the initial subjective trust value, refers to a quantitative index of the driver's initial trust level in the autonomous driving system without reference to historical data. For example, weights can be set according to the influencing degrees of the above-mentioned quantitative indexes such as machine uncertainty, machine transparency, and human conservatism value on the initial subjective trust value, and these indexes can be standardized to be in the same dimension and range; furthermore, the initial subjective trust value can be determined by weighted summation.

[0130] In this embodiment, the human subjective trust value is determined by combining the historical subjective trust value and the real-time performance score, that is, the trust level that the driver can assign to the autonomous driving system is determined based on the overall performance of the current vehicle during the entire driving process, rather than only based on the overtaking performance of this time, thereby reducing the contingency of the evaluation result of the human subjective trust value, enhancing its accuracy and reliability, and further enabling a more reasonable allocation of the human control weight during the vehicle driving process and reducing the risk of traffic accidents.

[0131] In a feasible implementation manner, step A20 includes:

[0132] Step A21, obtaining offline operation data, where the offline operation data includes the current trust value, the historical trust value, and the offline overtaking performance score;

[0133] Optionally, the offline operation data refers to a multi-dimensional data set collected through simulation or historical records in a non-real-time control environment, including, for a certain overtaking process, its corresponding offline overtaking performance score, the human subjective trust value (current trust value) determined after the overtaking is completed, and the historical trust value before the overtaking.

[0134] Step A22, fitting the current trust value, the historical trust value, and the offline overtaking performance score by the least squares method to determine the calculation weight coefficient of the human subjective trust value;

[0135] Optionally, the calculation weight coefficient refers to a parameter used to determine the relative importance of each index in the human subjective trust value, and can include a human personalization parameter for measuring the update rate of the historical subjective trust value and a weight for determining the influence of the real-time performance score on the human subjective trust value.

[0136] Exemplarily, for the current trust value historical trust value and based on the offline overtaking performance score and the machine objective ability value determined system trust value Combined with the following formula:

[0137]

[0138] Among them, represents the human personalization parameter, represents the influence weight of the real-time performance score on the human subjective trust value. The values of two intermediate parameters θ1 and θ2 can be calculated by using the least squares method based on the obtained offline operation data; and then, based on θ1 and θ2, the estimated value is determined.

[0139] Step A23: Determine the human subjective trust value according to the historical subjective trust value, the real-time performance score, and the calculated weight coefficient.

[0140] Exemplarily, after determining the human personalization parameter and the influence weight representing the influence of the real-time performance score on the human subjective trust value, the human subjective trust value can be determined with reference to the following formula

[0141]

[0142] Among them, represents the historical subjective trust value, represents the real-time performance score; is used to retain the previous trust value, while represents the update contribution of the current overtaking performance to the human subjective trust value.

[0143] Exemplarily, please refer to Figure 2 , Figure 2 provides a schematic diagram of the overall process for determining the human subjective trust value. It can be divided into three parts. The first part obtains the offline operation data by executing B101, and determines the calculation weight coefficient of the human subjective trust value by executing B102 according to the data including the current trust value, the historical trust value, and the offline overtaking performance score contained therein; the second part obtains the real-time overtaking data by executing B201, determines the index scores corresponding to the five indicators of the initial distance, the parallel boundary distance, the overtaking distance, the minimum distance, and the lateral acceleration according to the real-time overtaking data, and then executes B202 to determine the real-time performance score of the current vehicle according to the above five index scores; the third part obtains the historical subjective trust value by executing B301; finally, executes B110 to determine the final human subjective trust value according to the determined historical subjective trust value, the real-time performance score, and the calculated weight coefficient, and then this human subjective trust value can be used for subsequent steps such as human-machine trust value determination and control weight determination.

[0144] In this embodiment, by determining the calculation weight coefficient of the human subjective trust value through offline running data fitting, the relative importance of each index in the human subjective trust value can be determined more accurately, which is beneficial to improving the accuracy of the human subjective trust value determined during the actual driving of the vehicle, realizing a more accurate distribution of the human control weight in the automatic driving system, and further reducing the risk of traffic accidents when controlling the vehicle through the human-machine hybrid system.

[0145] Exemplarily, to help understand the implementation process of the vehicle control method based on human-machine trust obtained by combining the above Embodiment 1, please refer to Figure 3 , Figure 3 A general framework schematic diagram of a vehicle control method based on human-machine trust is provided. Specifically:

[0146] The preceding vehicle information module stores the relevant information of the vehicle being overtaken during the entire overtaking process, which is used to provide calculation data for the machine performance reference module. Furthermore, the machine performance reference module can determine indicators such as the initial distance between the current vehicle and the vehicle being overtaken, the minimum distance between the two, and the overtaking distance between the two after returning to the lane after overtaking completion during the overtaking process, and combine indicators such as the parallel boundary distance and lateral acceleration of the current vehicle to determine the real-time performance score of the current vehicle. Furthermore, the trust value calculation module can determine the human subjective trust value of the current vehicle after overtaking completion based on the obtained real-time performance score, in combination with parameters such as the calculation weight coefficient and historical subjective trust value stored in the system.

[0147] Furthermore, the control authority allocation module can determine the human control weight of the current vehicle based on this human subjective trust value, and combine the machine control input (machine control instruction) provided by the automatic driving module and the human control input (human control instruction) provided by the human driver module to determine the actual control input finally used to control the current vehicle to drive.

[0148] It should be noted that the above example is only for understanding the present application and does not constitute a limitation on the vehicle control method based on human-machine trust of the present application. Any simple transformation in more forms based on this technical concept is within the protection scope of the present application.

[0149] The embodiment of the present application also provides a vehicle control device based on human-machine trust. Please refer to Figure 4 , the vehicle control device based on human-machine trust includes:

[0150] The trust value determination module 10 is used to determine the human-machine trust value according to the obtained real-time overtaking data of the current vehicle;

[0151] The weight determination module 20 is used to determine the human control weight of the current vehicle according to the human-machine trust value;

[0152] A vehicle control module 30 is configured to control the current vehicle to travel based on the obtained human control instruction, machine control instruction, and human control weight.

[0153] The vehicle control device based on human-machine trust provided by the embodiments of the present application adopts the vehicle control method based on human-machine trust in the above embodiments, and can reduce the risk of traffic accidents when controlling a vehicle through a human-machine hybrid system. Compared with the prior art, the beneficial effects of the vehicle control device based on human-machine trust provided by the present application are the same as those of the vehicle control method based on human-machine trust provided by the above embodiments, and other technical features in the vehicle control device based on human-machine trust are the same as the features disclosed in the above embodiment method, and will not be elaborated herein.

[0154] The embodiments of the present application provide an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the vehicle control method based on human-machine trust in Embodiment 1 above.

[0155] Next, refer to Figure 5 , which shows a schematic structural diagram of an electronic device suitable for implementing the embodiments of the present application. The electronic device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions: tablet computers), PMPs (Portable Media Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The electronic device shown is only an example and should not impose any limitations on the functions and usage scopes of the embodiments of the present application.

[0156] As Figure 5As shown, the electronic device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which may perform various appropriate actions and processes according to the program stored in the read-only memory 1002 or the program loaded from the storage device 1003 into the random access memory 1004. In the random access memory 1004, various programs and data required for the operation of the electronic device are also stored. The processing device 1001, the read-only memory 1002, and the random access memory 1004 are connected to each other through a bus 1005. The input / output interface 1006 is also connected to the bus. Generally, the following systems may be connected to the input / output interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 may allow the electronic device to communicate with other devices wirelessly or wiredly to exchange data. Although the figure shows an electronic device having various systems, it should be understood that it is not required to implement or include all the shown systems. More or fewer systems may be implemented or included alternatively.

[0157] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from the network through the communication device, or installed from the storage device 1003, or installed from the read-only memory 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments disclosed in the present application are executed.

[0158] The electronic device provided by the embodiments of the present application adopts the vehicle control method based on human-machine trust in the above embodiments, and can reduce the risk of traffic accidents when controlling a vehicle through a human-machine hybrid system. Compared with the prior art, the beneficial effects of the electronic device provided by the present application are the same as those of the vehicle control method based on human-machine trust provided by the above embodiments, and other technical features in the electronic device are the same as those disclosed in the method of the previous embodiment, and will not be elaborated here.

[0159] It should be understood that the various parts disclosed in the present application may be implemented by hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0160] As described above, this is only the specific implementation of the present application. However, the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all should be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0161] The embodiments of the present application provide a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the vehicle control method based on human-machine trust in the above embodiments.

[0162] The computer-readable storage medium provided by the embodiments of the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program, and the program can be used by or combined with an instruction execution system or device. The program code contained on the computer-readable storage medium can be transmitted by any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0163] The above computer-readable storage medium may be included in an electronic device; or it may exist separately without being assembled into the electronic device.

[0164] The above computer-readable storage medium carries one or more programs. When the one or more programs are executed by an electronic device, the electronic device: determines a human-machine trust value according to the obtained real-time overtaking data of the current vehicle; determines the human control weight of the current vehicle according to the human-machine trust value; and controls the current vehicle to travel based on the obtained human control instructions, machine control instructions, and human control weight.

[0165] Computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The above-mentioned programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0166] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0167] The modules described in the embodiments of this application can be implemented in software or in hardware. Among them, the name of the module does not constitute a limitation on the unit itself in some cases.

[0168] The readable storage medium provided by the embodiments of this application is a computer-readable storage medium. The computer-readable storage medium stores computer-readable program instructions (i.e., computer programs) for performing the above-mentioned vehicle control method based on human-machine trust, which can reduce the risk of traffic accidents when controlling a vehicle through a human-machine hybrid system. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by this application are the same as those of the vehicle control method based on human-machine trust provided by the above embodiments, and will not be elaborated here.

[0169] An embodiment of the present application further provides a computer program product, including a computer program, which when executed by a processor implements the steps of the vehicle control method based on human-machine trust as described above.

[0170] The computer program product provided by the embodiment of the present application can reduce the risk of traffic accidents when controlling a vehicle through a human-machine hybrid system. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as those of the vehicle control method based on human-machine trust provided by the above embodiment, and will not be elaborated here.

[0171] The above are only partial embodiments of the present application, and do not limit the patent scope of the present application accordingly. Any equivalent structural transformation made by using the content of the specification and drawings of the present application under the technical concept of the present application, or direct / indirect application in other related technical fields, is included in the patent protection scope of the present application.

Claims

1. A vehicle control method based on human-machine trust, characterized in that The vehicle control method based on human-machine trust includes: Determining a human-machine trust value according to the obtained real-time overtaking data of the current vehicle; Determining a human control weight of the current vehicle according to the human-machine trust value; Controlling the current vehicle to travel based on the obtained human control instruction, machine control instruction, and the human control weight.

2. The vehicle control method based on human-machine trust according to claim 1, characterized in that The step of determining the human-machine trust value according to the obtained real-time overtaking data of the current vehicle includes: Determining a human subjective trust value according to the real-time overtaking data; Determining the human-machine trust value according to the human subjective trust value and a preset machine objective ability value.

3. The vehicle control method based on human-machine trust according to claim 2, wherein, The step of determining the human subjective trust value according to the real-time overtaking data includes: Determining a real-time performance score of the current vehicle according to the real-time overtaking data; Determining the human subjective trust value according to a historical subjective trust value and the real-time performance score.

4. The vehicle control method based on human-machine trust according to claim 3, wherein The step of determining the human subjective trust value according to the historical subjective trust value and the real-time performance score includes: Obtaining offline operation data, where the offline operation data includes a current trust value, a historical trust value, and an offline overtaking performance score; Fitting the current trust value, the historical trust value, and the offline overtaking performance score by the least squares method to determine a calculation weight coefficient of the human subjective trust value; Determining the human subjective trust value according to the historical subjective trust value, the real-time performance score, and the calculation weight coefficient.

5. The vehicle control method based on human-machine trust according to claim 3, wherein The method further includes: In the case of determining the human subjective trust value for the first time, determining the historical subjective trust value according to the obtained machine uncertainty, machine transparency, and human conservatism value.

6. The vehicle control method based on human-machine trust according to claim 3, characterized in that, The real-time overtaking data includes: an initial distance, a parallel boundary distance, a lateral acceleration, an overtaking distance, a minimum distance, an initial speed, and an initial acceleration. The step of determining the real-time performance score according to the real-time overtaking data includes: Determining a first index score of the initial distance according to the initial distance, the initial speed, the initial acceleration, a preset safety distance, and a reaction time; Determining a second index score of the parallel boundary distance according to the width of a one-way lane, the width of the current vehicle, and the parallel boundary distance; Determining a third index score of the overtaking distance according to the overtaking distance and the safety distance; Respectively determining a fourth index score of the minimum distance and a fifth index score of the lateral acceleration; Determining the real-time performance score according to the first index score, the second index score, the third index score, the fourth index score, and the fifth index score.

7. The vehicle control method based on human-machine trust according to claim 1, wherein The method further includes: In the case where the human-machine trust value is less than or equal to a first preset threshold, keeping the human control weight as a preset first weight value; In the case where the human-machine trust value is greater than the first preset threshold and less than a second preset threshold, determining the human control weight as a second weight value, where the second weight value has an inverse proportional relationship with the human-machine trust value; In the case where the human - machine trust value is greater than or equal to the second preset threshold, maintain the human control weight as a preset third weight value, where the first weight value is greater than the third weight value.

8. An electronic device, characterized in that, The device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the computer program is configured to implement the steps of the vehicle control method based on human - machine trust according to any one of claims 1 to 7.

9. A storage medium, characterized in that, The storage medium is a computer - readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the vehicle control method based on human - machine trust according to any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a computer program, and when the computer program is executed by a processor, it implements the steps of the vehicle control method based on human - machine trust according to any one of claims 1 to 7.

Images