Dynamic risk adaptation in vehicle

Abstract

Dynamic risk adaptation in a vehicle is provided. The disclosure relates to a method, a system, a vehicle, and a computer-readable storage medium, and a computer program product. The method includes obtaining a baseline risk model generated based on one or more risk-associated parameters. The method further includes obtaining sensor data from a sensor system of the vehicle. The method further includes estimating one or more risk values associated with the one or more risk-associated parameters based on the obtained sensor data, and generating an employed risk model based on the one or more estimated risk values. Further, the method includes providing the generated employed risk model to determine an acceptable risk model for formulating a driving policy for the ADS.

Description

Technical Field

[0001] The disclosed technology relates to methods and systems for determining the risk of a vehicle in various locations and traffic scenarios. Specifically, but not exclusively, aspects and embodiments of the disclosed technology relate to determining an acceptable risk model for formulating a driving strategy for an automated driving system ADS of a vehicle. Background Art

[0002] During the past few years, the number of research and development activities related to autonomous vehicles has surged, and many different approaches are being explored. More and more modern vehicles have advanced driver assistance systems (ADAS) for improving vehicle safety and more general road safety. ADAS, which can be represented by adaptive cruise control (ACC), collision avoidance systems, forward collision warnings, etc., are electronic systems that can help the driver of a vehicle while driving. Today, there is ongoing research and development in many technical fields related to driving automation, including both the fields of ADAS and autonomous driving (AD). In this article, ADAS and AD will be referred to under the general term autonomous driving system (ADS), which includes all different automation levels (in particular, level 3 to level 5) as defined, for example, by the SAE J3016 levels of driving automation (0 to 5).

[0003] In the near future, it is expected that the inclusion of AD functions or ADS solutions in new cars will grow rapidly. ADS can be interpreted as a complex combination of various components, which can be defined as a system that performs the perception, decision-making and operation of the vehicle by electronics and machinery instead of human drivers or in collaboration with human drivers and introduces unsupervised automation into road traffic. This includes the processing of vehicles, destinations and understanding of surrounding things. Although the automated system has control over the vehicle, it allows human operators to leave all or at least part of the responsibility to the system. ADS usually combines various sensors such as radar, lidar, sonar, camera, navigation system (e.g., GPS), odometer and / or inertial measurement unit (IMU) to perceive the surrounding things of the vehicle, and the advanced control system can interpret the perception information based on these sensors to identify appropriate navigation paths and obstacles, free space areas and / or related signs.

[0004] Ensuring the safety and performance of ADS is one of the key challenges in moving from a well-performing proof-of-concept status to a viable product for mass production and use on public roads. Since it is not feasible to perform methods such as statistical brute force verification of safe operation at the vehicle level (as in, for example, a report that states that 6 billion miles need to be driven without fatal failures in the United States to show that safety is comparable to that of human drivers), there is a need for more sophisticated methods in this field. Alternative methods may include formal methods, in which the ADS is proven to be safe given a certain set of assumptions about its operating environment. For example, methods have been proposed under the assumption of regular behavior traffic in the absence of any additional uncertainty. Similarly, another method assumes a certain behavior of the surrounding traffic, and if this is the case, the ADS can be proven to be safe. However, in many cases, the safety risks that ADS will face come from situations and unforeseen traffic scenarios that are not within the constraints of traffic rules or even conventional driving scenarios, but from so-called corner cases or edge cases. These situations may involve undetected VRUs (vulnerable road users) or VRUs that suddenly appear at the perception boundary or around the vehicle from an obscured state. Other instances of unusual events may include very aggressive driving from nearby traffic participants or adverse weather or road conditions that cause unforeseen consequences.

[0005] Therefore, there is an urgent need in the art for new and improved solutions for determining risk models that solve complex road traffic scenarios to ensure the safety and performance of vehicles equipped with ADS, and correspondingly for formulating driving strategies based on such risk models. Summary of the invention

[0006] The technology disclosed herein is intended to mitigate, alleviate or eliminate one or more of the above-identified defects and shortcomings in the prior art to solve various problems related to improving safety and enhancing the performance of ADS systems by dynamically and adaptively determining risk models. In addition, it is proposed to form an ADS driving strategy based on a determined acceptable risk model. In several aspects and embodiments of the present disclosure, a risk model can be understood as a model that expresses the level of risk in a given situation based on the driver's (ADS or human driver's) ability to respond appropriately to the development of possible traffic conditions. These risk models can be used to develop ADS that exhibit improved safety and performance measures.

[0007] Various aspects and embodiments of the disclosed invention are defined in the accompanying independent and dependent claims.In the present context, the term "exemplary" is to be understood as serving as an example, instance or illustration.

[0008] A first aspect of the disclosed technology includes a method. A method for determining an acceptable risk model to form a driving strategy for an automated driving system ADS is proposed.

[0009] The method includes obtaining a baseline risk model generated based on one or more risk-associated parameters. Further, the method includes obtaining sensor data from a sensor system of a vehicle, wherein the vehicle is a vehicle operated by a human driver or a vehicle operated by an ADS. The sensor data includes sensor data of a perception and / or manipulation system of the vehicle and sensor data associated with the driving behavior of the vehicle at a specific location and / or in a specific scenario. The method further includes estimating one or more risk values ​​associated with one or more risk-associated parameters based on the obtained sensor data. The method further includes generating an adopted risk model by updating one or more model parameters of a baseline risk model for a specific location and / or scenario based on the one or more estimated risk values. Further, the method includes providing the generated adopted risk model to determine an acceptable risk model for formulating a driving strategy for the ADS.

[0010] In several embodiments and aspects, a baseline risk model may be generated based on one or more risk-associated parameters. The risk-associated parameters may include any of the statistical probabilities of occurrence of various types of accidents or incidents in a specific location and / or a specific scenario, the statistical probabilities and / or properties of one or more external objects present in the surrounding environment of the vehicle in a specific location and / or a specific scenario, or the driving behaviors of multiple human driver-operated vehicles or multiple ADS-operated vehicles in a specific location and / or a specific scenario.

[0011] The driving behavior of a vehicle operated by a human driver may include at least the driver's reaction time, acceleration, braking, steering, lane keeping or speed control and / or the driver's gaze direction and trajectory of visual attention and / or a safety margin selected by the driver. The driving behavior of a vehicle operated by an ADS may include at least the vehicle's ADS-controlled acceleration, braking, steering, lane keeping or speed control.

[0012] In several embodiments, the specific location may include at least a specific geographic location and / or a specific type of location on a map. The specific location may include an intersection. The specific scene may include at least an occlusion state in the surrounding environment of the vehicle.

[0013] In several embodiments and aspects, the method may further include providing the generated adopted risk model to determine an acceptable risk model by transmitting the generated adopted risk model of the vehicle to a remote server for determining acceptable risk models based on multiple adopted risk models transmitted by vehicles operated by multiple human drivers or vehicles operated by multiple ADSs.

[0014] In several aspects and embodiments, the vehicle may include an automated driving system ADS. In several aspects and embodiments, the method may be performed by a control system of the vehicle including processing circuits configured to perform various embodiments and steps of the method. In several aspects and embodiments, the method may be performed by a control system of a server or data center including processing circuits configured to perform various embodiments and steps of the method.

[0015] In several aspects and embodiments, the vehicle is an ADS-equipped vehicle, wherein the method may further include receiving from a remote server a determined acceptable risk model formed based on integrated information obtained from multiple human driver-operated vehicles and multiple ADS-operated vehicles. The method may further include receiving from the remote server an updated driving strategy for the ADS formulated based on the determined acceptable risk model.

[0016] In several embodiments, the integrated information may include a generated adopted risk model for multiple human driver operated vehicles and multiple ADS operated vehicles transmitted to a remote server. The determined acceptable risk model may include one or more globally updated baseline risk model parameters determined based on the integrated information that meets the acceptable risk criteria.

[0017] In several embodiments and aspects, the acceptable risk criteria may include a risk of adoption of a human driver operated vehicle that indicates a safety improvement performance of the human driver operated vehicle.

[0018] The inventors have recognized that it would be very valuable to be able to dynamically adjust the acceptable risk of an ADS based on locations and scenarios where a human driver would accept a slightly higher level of risk. In any such case, the dynamic risk assumed by the ADS is always designed to be safe, but as a result of the variability, the performance of the ADS can be significantly improved within a range of driving behaviors that are still considered safe. It is further recognized that identifying where this may be possible can be accomplished by studying driver risk variability.

[0019] Executable instructions for performing these functions and embodiments of the method are optionally included in a non-transitory computer-readable storage medium or other computer program product configured to be executed by one or more processors.

[0020] According to a second aspect of the present disclosure, there is provided a (non-transitory) computer-readable storage medium comprising instructions which, when executed by one or more processors of a computing device of a vehicle, cause the computing device to perform a method according to any one of the embodiments of the method disclosed herein.

[0021] According to a third aspect of the present disclosure, there is provided a computer program product comprising instructions which, when the program is executed by one or more processors of a computing device of a vehicle, cause the computing device to perform a method according to any one of the embodiments of the method disclosed herein.

[0022] According to another fourth aspect, a system for determining an acceptable risk model to formulate a driving strategy for an autonomous driving system ADS is provided. The system includes a processing circuit configured to obtain a baseline risk model generated based on one or more risk-associated parameters. The processing circuit is further configured to obtain sensor data from a sensor system of a vehicle, wherein the vehicle is a vehicle operated by a human driver or a vehicle operated by an ADS. The sensor data includes sensor data of a perception and / or manipulation system of the vehicle and sensor data associated with the driving behavior of the vehicle at a specific location and / or in a specific scene. The processing circuit is further configured to estimate one or more risk values ​​associated with one or more risk-associated parameters based on the obtained sensor data. Further, the processing circuit is configured to generate an adopted risk model by updating one or more model parameters of a baseline risk model for a specific location and / or scene based on one or more estimated risk values. The processing circuit is further configured to provide a generated adopted risk model to determine an acceptable risk model for formulating a driving strategy for an ADS.

[0023] According to yet another fifth aspect, a vehicle is provided comprising one or more onboard sensors configured to monitor the surrounding environment, handling and / or driving behavior of the vehicle. The vehicle further comprises a system according to the fourth aspect and various embodiments of the fourth aspect. The vehicle may further comprise an ADS system for controlling one or more of the acceleration, steering and braking of the vehicle.

[0024] Further embodiments of the different aspects are defined in the dependent claims.

[0025] It should be noted that all embodiments, elements, features and advantages associated with the first aspect are also similarly applicable to the second, third, fourth and fifth aspects of the present disclosure.

[0026] These and other features and advantages of the present disclosure will be further elucidated in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above-described aspects, features, and advantages of the disclosed technology will be more fully understood by referring to the following illustrative and non-limiting detailed description of example embodiments of the present disclosure when taken in conjunction with the accompanying drawings. The drawings are not drawn to scale.

[0028] Figure 1A schematic top view of a road and at least one vehicle traveling on the road according to several embodiments of the present disclosure is illustrated.

[0029] Figure 2 A schematic top view showing multiple vehicles traveling on a road and a system according to several embodiments of the present disclosure.

[0030] Figure 3 Shown are schematic flow charts illustrating methods according to several embodiments of the present disclosure.

[0031] Figure 4 A schematic side view illustration of a vehicle including a control system according to some embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0032] The present disclosure will now be described in detail with reference to the accompanying drawings, in which some example embodiments of the disclosed technology are shown. However, the disclosed technology may be implemented in other forms and should not be construed as being limited to the disclosed example embodiments. The disclosed example embodiments are provided to fully convey the scope of the disclosed technology to the technician. It will be understood by those skilled in the art that the steps, services, and functions described herein may be implemented using separate hardware circuits, using software that works in conjunction with a programmed microprocessor or general-purpose computer, using one or more application-specific integrated circuits (ASICs), using one or more field programmable gate arrays (FPGAs), and / or using one or more digital signal processors (DSPs).

[0033] It will also be understood that when the present disclosure is described in terms of a method, it may also be implemented in a device comprising one or more processors, one or more memories coupled to the one or more processors, wherein computer code is loaded to implement the method. For example, in some embodiments, the one or more memories may store one or more computer programs that, when executed by the one or more processors, perform the steps, services, and functions disclosed herein.

[0034] It is also to be understood that the terms used herein are only used for the purpose of describing a particular embodiment and are not intended to be limiting. It should be noted that, as used in the specification and the appended claims, unless the context clearly specifies otherwise, the words "one", "an", "the" and "said" are intended to mean that there are one or more elements. Therefore, for example, in some contexts, references to "unit" or "the unit" may refer to more than one unit, etc. In addition, the words "include", "comprise" do not exclude other elements or steps. It should be emphasized that when used in this specification, the term "include" is used to indicate that there are stated features, wholes, steps or components. It does not exclude the presence or addition of one or more other features, wholes, steps, components or their groups. The term "and / or" should be interpreted as also meaning "both" and each is alternative. The term "obtain" is broadly interpreted in this article and includes receiving, retrieving, collecting, acquiring, etc. As used in this article, depending on the context, the term "if" can be interpreted as meaning "when..." or "when..." or "in an instance of..." or "in response to determining..." or "in response to detecting...". Similarly, depending on the context, the phrase “if it is determined that…” or “when it is determined that…” or “in an instance of…” may be interpreted to mean “upon determining that…” or “in response to determining that…” or “upon detecting and identifying the occurrence of an event” or “in response to detecting the occurrence of an event”.

[0035] It will also be understood that, although the terms first, second, etc. can be used to describe various elements or features in this article, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, without departing from the scope of the embodiment, the first signal can be referred to as the second signal, and similarly, the second signal can be referred to as the first signal. Both the first signal and the second signal are signals, but they are not the same signal.

[0036] In the following description of exemplary embodiments, the same reference numerals denote the same or similar components.

[0037] Ensuring the safety of ADS may include improved or at least comparable performance compared to human drivers. It should be noted that a huge contribution to the safety of the traffic environment is the extraordinary ability of humans to analyze challenging situations and also compensate for mistakes or unpredictable behavior of other drivers.

[0038] This can be achieved by ensuring that the ADS does not take any high risk at any location in its operating domain. However, in order to operate the vehicle in a reasonable manner and reach the destination, humans often accept higher risks under certain conditions. For example, driving through a four-way intersection will be associated with the risk of speeding vehicles or vehicles approaching dangerously from an orthogonal direction, which will be difficult to avoid safely from the side of the own vehicle. This is particularly important for traffic locations and scenarios involving visual impairment, blind spots or obstructed areas in the surrounding environment of the vehicle. For example, buildings that block the driver's field of vision of the vehicle and sensor devices installed on the vehicle. Despite this, human drivers often assess traffic conditions, make decisions, take various levels of risk, and accordingly enter intersections such as intersection 24a as described in more detail below at high and low speeds. Therefore, in order to achieve a safe and still effectively operating ADS, a variable level of risk acceptance should be considered because simply setting a constant risk limit or threshold across all operating conditions, scenarios and domains will result in obstructed and impaired ADS functions and performance.

[0039] However, it is not obvious from the prior art to understand what these risk levels should be and how they differ for human drivers and automated systems.The systems and methods described herein provide a solution for estimating the dynamic risk accepted by humans in various locations and traffic scenarios.

[0040] In the present context, risks may include potential risks associated with the likelihood of an event or scenario occurring in the vehicle's surroundings, such as a pedestrian walking out from behind a building or an obstructed space, a car making an unexpected turn, a building that is a school building with many children who may behave unpredictably, etc.

[0041] One objective herein is to identify and learn from such risk-prone scenarios having features or elements that make them more susceptible to potential dangers, uncertainties, or adverse outcomes in order to improve the behavior of AD systems.

[0042] Additionally or alternatively, risk may arise from a misunderstanding of a given scenario and the different ways in which a driver may perceive and react to a given scenario. For example, a driver may drive as if the road were empty, but ignore the presence of a pedestrian who suddenly walks into the road and in front of the vehicle. Thus, in such a scenario, there is a discrepancy between how the driver should behave (i.e., be more alert and careful) and how the driver actually behaves due to a misunderstanding of the dangerous scenario and a corresponding failure to react.

[0043] Embodiments and aspects herein provide solutions for analyzing various scenarios and events to identify sources of risk taken by a human driver or ADS operator, thereby improving the capabilities of an AD system.

[0044] Accordingly, the dynamic risk adopted by human drivers, which safely improves the performance and user experience of human-operated vehicles, is used as input to adaptively determine the quantitative safety and / or quality requirements of the ADS system, and then formulate the driving strategy of the ADS. When driving, humans usually accept risks differently depending on the environment in which they are driving. The risks accepted by humans can usually be implicit, and even the drivers themselves do not know, that is, there is no assigned quantified value. However, an ADS equipped with a risk assessment system such as a risk map or a certain risk model will be able to accurately assess the risks taken by humans when driving. It is known that the risk taken by humans is one of the potential factors leading to traffic accidents occurring today. Ideally, ADS is developed to exceed the driving behavior of human drivers in terms of accidents and safety, however, as mentioned above, a single risk metric that should be supported across all operating conditions and situations will lead to a very conservative execution of ADS. This will cause ADS to avoid manipulations that can be safely performed in quantity, and will make ADS more versatile.

[0045] The inventors have recognized that it would be very valuable to be able to dynamically adjust the acceptable risk of an ADS based on locations and scenarios where a human driver would accept a slightly higher level of risk. In any such case, the dynamic risk assumed by the ADS is always designed to be safe, but as a result of the variability, the performance of the ADS can be significantly improved within a range of driving behaviors that are still considered safe. It is further recognized that identifying where this may be possible can be accomplished by studying driver risk variability.

[0046] Figure 1 A schematic perspective view of an exemplary four-way intersection 24a is shown, depicting a traveling vehicle 1 on the four-way intersection 24a in a hatched pattern. The intersection 24a may be a portion of the portion 24 of the road 22. In addition, Figure 1At least one external vehicle 2 is shown traveling on the same portion 24 of the road 22 as the vehicle 1 but approaching an intersection 24a in a direction 2a transverse to the direction of travel 1a of the vehicle 1. In several examples and embodiments, the road 22 may be any type of road, such as a portion of a highway, an expressway, or an inner-city highway. The road may also be a country road, a rural road, a portion of a parking lot, or a portion of a garage facility, or any other suitable lane. The road may have multiple lanes such as more than one lane in the same direction of travel (e.g., two or more lanes 22a, 22b, 22c, 22d) or at least one lane in each direction of travel as is typically the case with rural roads. In this example, the road 22 has two lanes 22a and 22b extending in a longitudinal direction and two lanes 22c, 22d extending in a transverse or horizontal direction. The lanes 22a, 22b may include different types of lane markings, such as lane markings showing dashed lane marking types and lane markings showing solid lane marking types. In some examples, the road may lack any specific lane markings. In this example, vehicle 1 is traveling on lane 22b, and vehicle 2, approaching from the east, is traveling on lane 22c. The intersection has a building 30 located at its eastern corner, so that the view of vehicle 1 approaching intersection 24a is at least partially blocked. In the present context, the field of view or vision of vehicle 1 means the vision of the driver of vehicle 1 or the field of view of the vehicle's sensor device such as sensor system 320 including cameras, radars, lidars, etc. The field of view of vehicle 1 may be partially or completely blocked by obstacles such as building 30 in the example intersection 24a.

[0047] As described above, in several embodiments and aspects, the vehicle 1 includes a sensor system 320 configured to detect scenes that may occur in the surroundings of the vehicle 1. Therefore, the sensor system 320 is configured to collect information about the surroundings of the vehicle 1. The surroundings of the vehicle can be understood as a general area around the vehicle 1 in which scenes including objects (such as other vehicles, landmarks, obstacles, etc.) or places can be detected and identified by the vehicle sensors (radar, lidar, camera, etc.) (i.e., within the sensor range of the vehicle). The sensor system 320 may further include a positioning system for measuring various parameters associated with the vehicle, a speed and acceleration measurement sensor device, an eye tracking sensor device, an acceleration and / or brake pedal sensor device, etc. Therefore, the sensor system 320 may include a sensor device for recording and generating sensor data for the perception and / or manipulation system of the vehicle 1.

[0048] The vehicle 1 includes a control system 10, which may be otherwise referred to as a control device 10. The control system 10 of the vehicle 1 includes a control circuit 11, which may be otherwise referred to as a processing circuit 11, configured to obtain data including information about the surroundings of the vehicle 1, sensor data associated with a perception and / or manipulation system, etc. The information and sensor data of the surroundings of the vehicle 1 and the perception and / or manipulation system of the vehicle may be obtained from a sensor system 320 of the vehicle.

[0049] In some embodiments, the vehicle 1 may include a driving automation system. In this disclosure, in order to simply assist the reader, reference has been made to ADS, which includes both ADAS and AD systems. In some embodiments, the vehicle's ADS may include one or more ADS features, preferably level 2 features or higher features, according to the SAE J3016 level of driving automation for vehicles on the road. In the present context, the ADS features may be in the form of autonomous driving features, traffic jam pilot, highway pilot, or any other SAE J3016 level (such as level 1 and higher ADS features). In some embodiments, the control system 10 of the vehicle 1 may be a separate entity, or may be part of another control system such as the overall ADS architecture of the vehicle, and may accordingly be a module or component of the ADS.

[0050] In several aspects and embodiments, the vehicle 1 may include systems and modules (not specifically shown) as part of the vehicle's control system 10, such as a collision warning module, a steering module, a brake control module, a traction control module, etc. In several aspects and embodiments, these modules may be configured to obtain sensor data associated with the operation of the vehicle itself and information about the surrounding environment of the vehicle 1. In several embodiments, these modules may be configured to monitor and record events and traffic conditions that may occur on the road. For example, driver behavior including steering, braking, decisions about safety margins, the frequency or severity of potential collision events identified or predicted, emergency braking, and other risk-associated scenarios may be recorded and entered into a formal record. In an ADS-equipped vehicle, the driving behavior of the ADS and corresponding events such as braking, steering, predicted or identified collision events, etc. may be recorded for such ADS-controlled functions.

[0051] The acquired sensor data from the sensor system 320 may also include information about at least one stationary or moving object in the vehicle's surroundings. Figure 1As depicted in FIG. 1 , in addition to the external vehicle 2, there is at least one object 4 in the surroundings of the vehicle 1. The at least one object 4 may be a VRU such as an animal, a wild animal (not shown), a pedestrian 4 that appears on the lane 22a and attempts to cross the road 22 in the direction of the arrow 4a. At least one external object 4 and / or external vehicle 2 may appear on the road 22 on specific occasions, and may appear in the surroundings of the vehicle at a specific location and / or in a specific scenario.

[0052] In addition, the obtained sensor data may include information about lane markings for one or more lanes of the road 22, such as lane marking types (e.g., solid lines, dashed lines, double markings, etc.), traffic signs or signal information, road obstacle information such as the presence or characteristics of speed bumps or any other road obstacles, etc. In several embodiments, the control system 10 can access map data for identifying specific locations on the road. The map data may include high-definition (HD) map data. Therefore, any specific location in the current context may include at least a specific geographic location on the map and / or a specific type of location including, for example, an intersection or a bridge or a location associated with a landmark, road obstacle, etc. The specific scene in the current context may include at least a specific scene such as a Figure 1 In the example of FIG. 3 , the occlusion state of the vehicle's surroundings of the building 30 is shown. Additionally or alternatively, any other traffic scenarios such as cruising, overtaking, parking, reversing, etc. are conceivable.

[0053] In several embodiments and examples, the sensor data may be obtained at least in part in real time from the sensor system 320. In some exemplary embodiments, the sensor data may be obtained at least in part from the memory unit 12 of the vehicle 1. Accordingly, images captured by the camera sensor and / or lidar and / or radar data may be stored locally in the memory 12 of the vehicle 1 and / or transmitted to an external network 20 such as a cloud network 20 or to a remote server 15 with which the vehicle 1 may communicate.

[0054] Various aspects and embodiments of the proposed technology and corresponding systems and methods may be implemented and executed by a processing circuit 11 including one or more processors. The processing circuit 11 may be at least partially included in a control system 10 or device implemented in the vehicle 1. In some embodiments, the control system and processing circuit may be a decentralized system and implemented in a remote server 15 in communication with the vehicle 1.

[0055] Accordingly, the control system 10 of the vehicle 1 may be configured to register and determine the occurrence of scenes in the surroundings of the vehicle 1. Scenes in some embodiments may correspond to locations or events such as accidents that may occur in the surroundings of the vehicle 1. In some embodiments, a scene may be the presence of at least one object 4 or external vehicle 2 in the surroundings of the vehicle 1 determined based on the obtained sensor data. Scenes may also be associated with specific geographic locations such as the intersection 24a, specific environmental parameters such as weather conditions, time parameters such as time of day, rush hour traffic, school zone speed limits at specific times, etc. Scenes may also be affected by various lighting conditions including daylight, darkness, or nighttime lighting on a road or at a specific location such as the intersection 24a. Specific scenes and / or specific locations may be associated with their respective risk parameters.

[0056] For example, in Figure 1 In the figure, two possible scenarios corresponding to potential risk events have been depicted by way of examples.

[0057] In the first scenario, an external vehicle 2 traveling on lane 22c is approaching an intersection 24a in a direction 2a transverse to the direction of travel 1a of vehicle 1. The field of view of vehicle 1 approaching intersection 24a is at least partially blocked by building 30. The scenario involves the probability of vehicle 2 actually approaching intersection 24a. Therefore, vehicle 1 may be prone to a potential collision event with vehicle 2, and the possibility of collision has one or more risk parameters associated with the driving behavior of vehicle 1. In the case where vehicle 1 is operated by ADS, how the human driver or ADS controller of vehicle 1 estimates the probability of vehicle 2 entering intersection 24a and perceives the risk of collision with vehicle 2 and acts accordingly will determine how vehicle 1 safely and effectively drives through intersection 24a. As described above, the risk can be partially affected by the probability of vehicle 2 emerging from the obstructed space behind building 30 and the model of the speed, acceleration, etc. of the vehicle 1 when approaching intersection 24a.

[0058] In the second scenario, a VRU (i.e., pedestrian 4) may appear in front of vehicle 1, crossing the road on lane 22a, causing a safety issue for vehicle 1. Furthermore, depending on how vehicle 1 behaves with respect to the first scenario (i.e., speed, acceleration, braking, deceleration, acceleration, etc.) when observing vehicle 2 approaching from the east, a new traffic scenario with its own associated risk parameters may arise with respect to pedestrian 4.

[0059] In several aspects and embodiments, sensor data including sensor data of the perception and / or manipulation system of the vehicle 1 and sensor data associated with the driving behavior of the vehicle at a specific location 24a and / or in a specific first scenario and / or second scenario will be recorded and processed by the control system 10. Accordingly, the risk of adoption can be quantified for the vehicle 1, for example, based on the trajectory and / or distance data of the pedestrian 4 and / or the relative speed data of the vehicles 1 and 2 perceived by the driver and / or obtained from the sensor system 320.

[0060] In any of these scenarios, the human driver and the ADS controller may adopt and accept different risk levels. For example, the human driver may be aware of potential road congestion or temporary road maintenance operations on road 22 and lanes 22c and 22d. In another example, the driver may be aware that intersection 24a is located in a remote area or building 30 is an abandoned building without any residents and therefore other vehicles or VRUs are rarely present in the area.

[0061] Thus, even though the building 30 is blocking the driver's view, the human driver can estimate a very low or near-zero probability that another vehicle, such as vehicle 2, is approaching from the east. Thus, the driver of vehicle 1 can drive through the intersection at a higher speed than if the ADS controller had conservatively slowed down regardless of the fact that no other vehicles 2 or VRUs might enter the intersection 24a. In the absence of pedestrians 4, the higher speed of the driver-operated vehicle 1 across the intersection 24a can be considered an efficient behavior for safe operation with a low risk of perception by the driver. Similarly, the driving behavior of the ADS controller can be recorded as an inefficient behavior with a high risk characteristic of adoption.

[0062] However, if a pedestrian 4 suddenly appears on that side of the intersection, intending to cross the road, the same behavior of the driver-operated vehicle 1 can be regarded as a high-risk and unsafe event that may result in a collision event with the pedestrian 4. On the other hand, the driving behavior of the ADS controller can be regarded as a high-risk but safe and effective driving behavior.

[0063] Similarly, simply observing the crosswalk 4b by the ADS controller may result in an unnecessary speed reduction of the vehicle 1 that perceives a high risk scenario and a high probability that a pedestrian 4 is always present on the road. However, the human driver may realize that the area is a school zone and at a certain time (e.g., at night), there are not many pedestrians present in the area, so approaching the intersection 24a at a higher speed may be a low risk but efficient behavior.

[0064] In several embodiments and aspects, the nature of the risk may be different for a human and an ADS controller. For example, approaching the intersection 24a at night and in dark conditions may be considered a high-risk scenario for a human driver with impaired night vision, while the same scenario may be considered a low-risk scenario for an ADS controller using cameras, radars, and lidar sensors for detecting objects on the road. Thus, an ADS controller may navigate the intersection 24a at a higher speed than a human driver while still being within the bounds of safe traffic behavior.

[0065] It should be clear to those skilled in the art that the risk scenarios depicted and described are simplified examples used only to illustrate the inventive systems and methods herein and may not fully represent the accurate real environment associated with these scenarios. Those skilled in the art also understand that for the purposes of this disclosure, many other scenarios that arise may be included in different embodiments and aspects and are equally considered.

[0066] In all of the above exemplary scenarios, the control system 10 will accordingly obtain sensor data associated with the operation and driving behavior of the vehicle 1 operated by a human driver or an ADS to quantify the risk adopted and accepted by the driver or ADS controller.

[0067] Repeatedly assessing the risk that humans accept in such situations in a group of individuals will provide a distribution of risk acceptance. This distribution can also provide valuable insights into the main causes of traffic accidents, which will be used for improvement when developing ADS and its driving strategy. Therefore, by dynamically adjusting the risk acceptable to the ADS in the same situation, scene and / or location so that it is less than the estimated distribution, a safer and more effective driving strategy for the ADS can be formulated.

[0068] In more detail, aspects and embodiments of the present disclosure provide a solution for determining an acceptable risk model to form a driving strategy for an ADS. The dynamically and adaptively adjusted acceptable risk model and ADS driving strategy can then be used to improve the safety and performance of vehicles equipped with an ADS.

[0069] To this end, the control system 10 of the exemplary vehicle 1 may be configured to obtain a baseline risk model that is typically generated based on one or more risk-associated parameters.

[0070] In some embodiments, a baseline risk model may be generated based on one or more risk-related parameters. The one or more risk-related parameters may include, for example, Figure 1The statistical probability of various types of accidents or events occurring at a specific location and / or a specific scenario discussed at the intersection 24a of the vehicle. Additionally or alternatively, one or more risk-associated parameters may include the statistical probability and / or attributes of one or more external objects (e.g., vehicle 2 or VRU 4) that appear in the vehicle's surroundings at a specific location and / or a specific scenario. Further, one or more risk-associated parameters may include the driving behavior of multiple human driver-operated vehicles or multiple ADS-operated vehicles at a specific location and / or a specific scenario. In other words, data associated with traffic scenarios and locations can be collected from a fleet of vehicles operated by human drivers or ADS-operated vehicles and used to build a baseline risk model. Vehicle 1 can obtain the baseline risk model from a remote server 15, or the baseline risk model can be manually uploaded to the memory 12 of the control system 10 of vehicle 1 at a maintenance or test station.

[0071] In several embodiments, the control system 10 can be configured to obtain sensor data from the sensor system 320 of the vehicle 1, wherein the vehicle can be a human driver-operated vehicle or an ADS-operated vehicle as described above. It should be understood that the vehicle operated by the human driver can be ADS-enabled (without ADS installed, but with the structure required to install ADS) or equipped with ADS, but operated by a human driver when recording the traffic scenarios and risk estimation functions described in this article. However, the human-operated vehicle 1 includes the sensor system 320 and the control system 10 as well as all other components and modules required to record and quantify the driver's risks of adoption in traffic conditions and geographic locations. As described above, the sensor data may include sensor data of the vehicle's perception and / or manipulation system and sensor data associated with the vehicle's driving behavior in a specific location and / or specific scenario.

[0072] The control system 10 can be configured to estimate one or more risk values ​​associated with one or more risk-associated parameters based on the obtained sensor data, and further generate an adopted risk model by updating one or more model parameters of a baseline risk model for a specific location and / or scene. The adopted risk model of the vehicle 1 can be updated based on one or more estimated risk values. In several embodiments, the driving behavior of a vehicle operated by a human driver can include at least the driver's reaction time, acceleration, braking, steering, lane keeping or speed control. Additionally or alternatively, the driver's gaze direction and the trajectory of visual attention can be monitored by an eye tracking sensor device and recorded as driver behavior. Further, the safety margin selected by the driver, the braking mode, the acceleration mode, the driver's actuation and / or release of the brake and / or accelerator pedal can all be recorded as the driving behavior of the human driver.

[0073] Similarly, in several embodiments, the driving behavior of the ADS-operated vehicle may include at least the vehicle's ADS-controlled acceleration, braking, steering, lane keeping, or speed control. Obviously, other functions of the vehicle controlled by the ADS may also be recorded and utilized.

[0074] In several embodiments herein, the generated adopted risk model may be provided by the control system 10 to determine an acceptable risk model and formulate a driving strategy for the ADS. By providing the generated risk model in the current context, it is meant that the control system 10 may be configured to provide the generated adopted risk model of the vehicle 1 to the external server 15 in real time (i.e., online). Additionally or alternatively, the control system 10 may be configured to provide the generated adopted risk model in an offline mode, i.e., the values ​​and parameters of the risk model may be stored in the memory 12 of the control system 1. The stored risk model may be provided to the remote server 15 at a later time (e.g., at a predetermined time interval or on a predetermined schedule such as once a day, once a week, etc.). The control system may be further configured to provide the generated adopted risk model to the server 15 on demand (i.e., upon receiving a request from the remote server 15). Accordingly, in several embodiments and aspects, providing the generated adopted risk model to determine the acceptable risk model may include transmitting the generated adopted risk model of the vehicle to the remote server 15 to determine the acceptable risk model based on a plurality of adopted risk models transmitted by vehicles operated by a plurality of human drivers or vehicles operated by a plurality of ADSs. In other words, the acceptable risk model can be based on Figure 2 Accordingly, one or more fleet vehicles may be connected to the external network 20 via a wireless link, for example, for transmitting risk model data and / or receiving queries from the remote server 15.

[0075] Among a plurality of vehicles in a convoy appearing on a road (e.g., traveling in a particular geographic location such as a geographic sub-area 200a including an intersection 24a), at least one subset 110 of vehicles may be human-operated, and another subset 120 of vehicles may be ADS-operated. The vehicles included in either subset 110, 120 are assumed to be equipped with a sensor system and a control system for performing at least some of the embodiments of the present disclosure.

[0076] For example, a subset of vehicles 110 shown as hatched vehicles may be human operated, and another subset 120 may be ADS operated. It should be appreciated that the fleet of vehicles may include any number of vehicles other than the hatched illustrated vehicles 110, and Figure 2The illustration of the example vehicles and traffic scenes in is merely to assist the reader. In addition, the vehicle subsets 110, 120 may appear at a geographic location such as the intersection 24a at the same time, or cross the intersection 24a at different points in time with associated time stamps. For example, where the intersection is included in a school zone or the like, the driving behavior of human-operated vehicles 110 or ADS-operated vehicles 120 at the same intersection 24a may be recorded for both daytime and nighttime, for example to investigate different behavior patterns of drivers during school hours.

[0077] The present inventors have recognized that by utilizing vehicle data from a fleet of vehicles, systems and methods can be provided for forming statistically relevant risk assessments of humans and ADS controllers in various locations and scenarios. Accordingly, analyzing the collected data can reveal patterns of behavior associated with various risk levels similarly employed by humans and ADS controllers. Thus, deep insights are provided for updating the acceptable risk model of the ADS controller based on one or more risk parameters employed by the human, which will lead to safely improving performance and updating the driving strategy of the ADS.

[0078] By frequently querying the risk model adopted by the vehicle fleet, the acceptable risk model of the ADS can be continuously updated and improved due to the large amount of data continuously generated by the fleet vehicles. In several embodiments and aspects, the acceptable risk model can be manually determined at the server 15, for example, by a human operator examining or analyzing the generated adopted risk model obtained from the fleet. In some exemplary embodiments, the server and processing circuit of the server 15 can be adapted to determine the acceptable risk model parameters at least partially automatically or based on input received from a human operator. The human operator can manually identify the acceptable risk level among the risks adopted by the human driver associated with the risky scene, and determine the safety and performance factors and parameters based on the characteristics of the identified scene or associated location. This approach ensures that the seemingly elevated risk parameters adopted by the human driver, which lead to improved performance and user experience, will be carefully examined for safety requirements. Therefore, the acceptable risk model parameters of the ADS will only be updated with the parameters of the risk model adopted by the driver who passes the safety requirements. The safety requirements can be acceptable risk standards defined for identifying and processing risk parameters associated with unstable, dangerous, irresponsible or aggressive driving behaviors and risky patterns of human drivers.

[0079] In several embodiments and aspects, a determined acceptable risk model may be formed based on integrated information obtained from multiple human driver operated vehicles and multiple ADS operated vehicles. The integrated information may include a statistical model of risk and performance levels formed based on an adopted risk model obtained from a fleet of vehicles.

[0080] In some embodiments, such as Figure 2 As shown in the block diagram of , the integrated information may include the generated adopted risk models 210, 220 of multiple human driver operated vehicles and multiple ADS operated vehicles transmitted to the remote server 15. The determined acceptable risk model may include one or more globally updated baseline risk model parameters determined based on the integrated information that meets the acceptable risk criteria. As described above, the acceptable risk criteria may include the adopted risk of human driver operated vehicles that indicates the safely improved performance of human driver operated vehicles in specific traffic scenarios and / or locations. In other words, the acceptable risk criteria may include the adopted risk of human driver operated vehicles that indicates the improved performance of human driver operated vehicles when they are still considered and determined to be safe.

[0081] The integrated information of the subset 110 can be compared 230 with the integrated information obtained from the subset 120 of the fleet for a particular scenario and / or location. In this way, differences in the risk level and resulting performance of the subset 110 operated by humans and the subset operated by ADS can be advantageously identified. After analysis and determination in the server 15, the ADS acceptable risk model parameters can be updated 240 to include similarity to human drivers, for example, in reference to Figure 1 The new knowledge can be used more effectively and safely to operate the vehicle in the example of the abandoned building 30 discussed. In this way, the overly cautious risk parameter adopted by the ADS can be updated with the knowledge that the probability of another vehicle 2 approaching from the east side of the intersection is extremely low, and the ADS-operated vehicle can therefore continue to travel by maintaining speed without unnecessary deceleration. Another possible result is that the ADS-operated vehicle slows down, but not as much as before the deceleration, depending on how large the updated risk perception is perceived to be.

[0082] In several embodiments, the driving strategy of the ADS may be developed and updated 250 in a remote server based on the determined acceptable risk model.

[0083] In such Figure 2 In several embodiments shown in the example of , the ADS-equipped vehicle 120 can receive from a remote server a determined acceptable risk model formed based on integrated information obtained from a fleet (i.e., multiple human driver-operated vehicles and multiple ADS-operated vehicles). In addition, the ADS-equipped vehicle 120 can receive from the remote server an updated driving strategy of the ADS formulated based on the determined acceptable risk model.

[0084] Figure 3 Various aspects and embodiments according to the present disclosure are shown and reference is made to Figure 1 and Figure 2Flowchart of method 400 of the technique presented in .

[0085] A method 400 for determining an acceptable risk model to formulate a driving strategy for an automated driving system ADS is proposed. The method 400 includes obtaining 401 a baseline risk model generated based on one or more risk-associated parameters. The method 400 further includes obtaining 403 sensor data from a sensor system 320 of a vehicle 1, wherein the vehicle is a vehicle operated by a human driver or a vehicle operated by an ADS. The sensor data includes sensor data of a perception and / or manipulation system of the vehicle and sensor data associated with the driving behavior of the vehicle at a specific location and / or in a specific scene. The method 400 further includes estimating 405 one or more risk values ​​associated with one or more risk-associated parameters based on the obtained sensor data, and generating 407 an adopted risk model by updating one or more model parameters of a baseline risk model for a specific location and / or scene based on the one or more estimated risk values. In addition, the method 400 includes providing 409 the generated adopted risk model to determine an acceptable risk model for formulating a driving strategy for the ADS.

[0086] In several aspects and embodiments, a baseline risk model may be generated based on one or more risk-associated parameters, including: any one of the statistical probabilities of various types of accidents or incidents occurring at a specific location and / or in a specific scenario; the statistical probability and / or attributes of one or more external objects appearing in the vehicle's surroundings at a specific location and / or in a specific scenario; or the driving behavior of vehicles operated by multiple human drivers or vehicles operated by multiple ADSs at a specific location and / or in a specific scenario.

[0087] In several embodiments, the driving behavior of a vehicle operated by a human driver may include at least the driver's reaction time, acceleration, braking, steering, lane keeping or speed control and / or the driver's gaze direction and trajectory of visual attention and / or a safety margin selected by the driver. The driving behavior of a vehicle operated by an ADS may include at least the acceleration, braking, steering, lane keeping or speed control of the vehicle 1 controlled by the ADS.

[0088] In several embodiments, the specific location may include at least a specific geographic location on a map and / or a specific type of location including an intersection. The specific scene may include at least an occlusion state in the surrounding environment of the vehicle.

[0089] In several embodiments, the method 400 may further include providing 409 the generated adopted risk model by transmitting 411 the generated adopted risk model of the vehicle to the remote server 15 to determine an acceptable risk model. Figure 2Illustratively, the acceptable risk model may be determined based on a plurality of adopted risk models transmitted by a plurality of human driver operated vehicles 110 or a plurality of ADS operated vehicles 120 .

[0090] In several embodiments herein, the vehicle may be an ADS-equipped vehicle 1, 120. The method 400 may further include receiving 413 from the remote server 15 a determined acceptable risk model formed based on integrated information obtained from multiple human driver-operated vehicles 110 and multiple ADS-operated vehicles 120. The method 400 may further include receiving 415 from the remote server an updated driving strategy of the ADS formed based on the determined acceptable risk model.

[0091] In several embodiments, the integrated information may include a generated adopted risk model for multiple human driver operated vehicles and multiple ADS operated vehicles transmitted to a remote server. The determined acceptable risk model may include one or more globally updated baseline risk model parameters determined based on the integrated information that meets the acceptable risk criteria. In several embodiments, the acceptable risk criteria may include the risk of adoption of a human driver operated vehicle that indicates a safe improved performance of a human driver operated vehicle. In other words, the risk parameters used to update the acceptable risk model may be risk parameters associated with the risk of adoption of a human driver that safely improved the performance of the vehicle. For example, the acceptable risk criteria may not include a risk level of erratic or dangerous risk taken by a human driver and its parameters, such as as described in reference to Figure 1 The illustrated scenario is when a human-operated vehicle 1 enters an intersection 24a at high speed and a pedestrian 4 appears on the roadside.

[0092] It should be noted that all embodiments, elements, features, examples and advantages previously described with reference to the control system 10 of the vehicle 1 or the fleet of vehicles 110 , 120 or the processing circuit 11 and the remote server 15 of the control system 10 are similarly and equally applicable to the various embodiments of the method 400 .

[0093] Executable instructions for performing these functions and embodiments of method 400 are optionally included in a non-transitory computer-readable storage medium or other computer program product configured to be executed by one or more processors.

[0094] Figure 41 is a schematic illustration of a vehicle 1 including a control system 10, which may also be referred to as a control device or apparatus 10 or simply as a device 10. As used herein, a "vehicle" is any form of motor vehicle. For example, the vehicle 1 may be any road vehicle, such as a car (as illustrated herein), a motorcycle, a (freight) truck, a bus, etc. The vehicle 1 may be an electric vehicle (EV) or a hybrid vehicle.

[0095] The device 10 includes a control circuit 11 and a memory 12. The control circuit 11 may physically include a single circuit device. Alternatively, the control circuit 11 may be distributed over several circuit devices. As an example, the device 10 may share its control circuit 11 with other parts of the vehicle 1 (e.g., the ADS 310). In addition, the control system 10 may form a part of the ADS 310, that is, the control system 10 may be implemented as a module or feature of the ADS. The control circuit 11 may include one or more processors such as a central processing unit (CPU), a microcontroller, or a microprocessor. One or more processors may be configured to execute program codes stored in the memory 12 so as to perform various functions and operations of the vehicle 1 in addition to the methods disclosed herein. The processor may be or include any number of hardware components for performing data or signal processing or for executing computer codes stored in the memory 12. Optionally, the memory 12 includes a high-speed random access memory, such as a DRAM, SRAM, DDR RAM, or other random access solid-state memory device; and optionally includes a non-volatile memory such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or other non-volatile solid-state storage devices. Memory 12 may include a database component, an object code component, a script component, or any other type of information structure for supporting the various actions of the present specification.

[0096] The ADS may access map data, for example, used by the ADS 310 of the vehicle 1 in order to perform autonomous functions of the vehicle 1 when implemented. The map data may include high-definition (HD) map data. It is contemplated that, although illustrated as a separate element from the ADS 310, the memory 12 may be provided as an integrated element of the ADS 310. In other words, according to an exemplary embodiment, any distributed or local memory device may be utilized when implementing the inventive concept. Similarly, the control circuit 11 may be distributed, for example, so that one or more processors of the control circuit 11 are provided as an integrated element of the ADS 310 of the vehicle 1 or any other system. In other words, according to an exemplary embodiment, any distributed or local control circuit device may be utilized when implementing the inventive concept. The ADS 310 is configured to perform functions and operations of autonomous or semi-autonomous functions of the vehicle 1. The ADS 310 may include a plurality of modules, wherein the task of each module is a different function of the ADS 310.

[0097] The vehicle 1 includes a number of elements commonly found in autonomous or semi-autonomous vehicles. It will be appreciated that the vehicle 1 can have Figure 4 In addition, the vehicle 1 may include any combination of the various elements shown in FIG. Figure 4 Although various elements are shown herein as being located inside the vehicle 1, one or more of the elements can be located outside the vehicle 1. For example, map data can be stored in a remote server 15 and accessed by various components of the vehicle 1 via the communication system 326. Furthermore, as will be readily appreciated by those skilled in the art, even though various elements are depicted herein in a certain arrangement, the various elements can be implemented in a different arrangement. It should also be noted that the various elements can be communicatively connected to each other in any suitable manner. Since the elements of the vehicle 1 can be implemented in a number of different ways, Figure 4 The vehicle 1 should be regarded as an illustrative example only.

[0098] The vehicle 1 further comprises a sensor system 320. The sensor system 320 is configured to obtain sensory data about the vehicle itself or its surroundings. The sensor system 320 may, for example, comprise a global navigation satellite system (GNSS) module 322 (such as GPS) configured to collect geographic location data of the vehicle 1. The sensor system 320 may further comprise one or more sensors 324. The sensor 324 may be any type of on-board sensor such as a camera, a lidar and a radar, an ultrasonic sensor, a gyroscope, an accelerometer, an odometer, etc. It should be understood that the sensor system 320 may also provide the possibility of obtaining sensory data directly or via a dedicated sensor control circuit in the vehicle 1.

[0099] The vehicle 1 further includes a communication system 326. The communication system 326 is configured to communicate with external units such as other vehicles (i.e., via a vehicle-to-vehicle (V2V) communication protocol), a remote server (e.g., a cloud server), a database, or other external devices (i.e., a vehicle-to-infrastructure (V2I) or a vehicle-to-everything (V2X) communication protocol). The communication system 326 can communicate using one or more communication technologies. The communication system 326 may include one or more antennas (not shown). Cellular communication technology can be used for remote communications such as to a remote server or a cloud computing system. In addition, if the cellular communication technology used has low latency, it can also be used for V2V communication, V2I communication, or V2X communication. Examples of cellular radio technologies are GSM, GPRS, EDGE, LTE, 5G, 5G NR, etc., also including future cellular solutions. However, in some solutions, medium- and short-range communication technologies such as wireless local area networks (LANs) for communicating with other vehicles in the vicinity of the vehicle 1 or with local infrastructure elements, such as solutions based on IEEE 802.11, can be used. ETSI is developing cellular standards for vehicle communications, and 5G, for example, is considered a suitable solution due to its low latency and efficient handling of high bandwidth and communication channels.

[0100] The communication system 326 may accordingly provide the possibility to send outputs to a remote location (e.g. a remote operator or a control center of a remote server 15) and / or receive inputs from a remote location via one or more antennas. In addition, the communication system 326 may be further configured to allow the various elements of the vehicle 1 to communicate with each other. As an example, the communication system may provide a local network setting such as CAN bus, I2C, Ethernet, fiber optics, etc. The local communication within the vehicle may also be of wireless type with protocols such as WiFi, LoRa, Zigbee, Bluetooth or similar medium / short range technologies.

[0101] The vehicle 1 further includes a steering system 328. The steering system 328 is configured to control the steering of the vehicle 1. The steering system 328 includes a steering module 330 configured to control the direction of the vehicle 1. The steering system 328 further includes a throttle module 332 configured to control the actuation of the throttle of the vehicle 1. The steering system 328 further includes a brake module 334 configured to control the actuation of the brakes of the vehicle 1. The various modules of the steering system 328 can also receive manual inputs from the driver of the vehicle 1 (i.e., from the steering wheel, accelerator pedal, and brake pedal, respectively). However, the steering system 328 can be communicatively connected to the vehicle's ADS 310 to receive instructions on how the various modules of the steering system 328 should act. Therefore, the ADS 310 is able to control the steering of the vehicle 1, for example, via the decision and control module 318.

[0102] The ADS 310 may include a positioning module 312 or positioning block / system. The positioning module 312 is configured to determine and / or monitor the geographic location and orientation of the vehicle 1 and may utilize data from a sensor system 320, such as data from a GNSS module 322. Alternatively or in combination, the positioning module 312 may utilize data from one or more sensors 324. Alternatively, the positioning system may be implemented as a real-time kinematic (RTK) GPS to improve accuracy.

[0103] The ADS 310 may further include a perception module 314 or perception block / system 314. The perception module 314 may refer to any well-known module and / or function, such as included in one or more electronic control modules and / or nodes of the vehicle 1, adapted and / or configured to interpret perception data related to driving of the vehicle 1 to identify, for example, obstacles, vehicle lanes, relevant signs, appropriate navigation paths, etc. Thus, the perception module 314 may be adapted to rely on and obtain input from multiple data sources such as automotive imaging, image processing, computer vision, and / or in-vehicle networking, etc., combined with, for example, perception data from the sensor system 320.

[0104] The positioning module 312 and / or the perception module 314 may be communicatively connected to the sensor system 320 to receive perception data from the sensor system 320. The positioning module 312 and / or the perception module 314 may also send control instructions to the sensor system 320. The ADS 310 may also include other modules such as a path planning module 316. In some embodiments, the positioning module 312 and / or the perception module 314 may be included in the sensor system 320, and vice versa.

[0105] The present invention has been presented above with reference to specific embodiments. However, other embodiments than those described above are possible and within the scope of the present invention. Within the scope of the present invention, method steps different from the above method steps may be provided for performing the method by hardware or software. Therefore, according to an exemplary embodiment, a (non-transitory) computer-readable storage medium is provided, which stores one or more programs, the one or more programs being configured to be executed by one or more processors of a vehicle control system, the one or more programs including instructions for performing a method according to any of the embodiments discussed above. Alternatively, according to another exemplary embodiment, a cloud computing system may be configured to perform any of the methods presented herein. The cloud computing system may include distributed cloud computing resources that jointly perform the methods presented herein under the control of one or more computer program products.

[0106] In general, a computer-accessible medium may include any tangible or non-transitory storage medium or memory medium, such as an electrical, magnetic, or optical medium, for example, a disk or CD / DVD-ROM connected to a computer system via a bus. As used herein, the terms "tangible" and "non-transitory" are intended to describe a computer-readable storage medium (or "memory") that does not include propagating electromagnetic signals, but are not intended to otherwise limit the types of physical computer-readable storage devices encompassed by the term computer-readable medium or memory. For example, the term "non-transitory computer-readable medium" or "tangible memory" is intended to cover types of storage devices that do not necessarily store information permanently, including, for example, random access memory (RAM). Program instructions and data stored in a non-transitory form on a tangible computer-accessible storage medium may be further transmitted via a transmission medium or signals, such as electrical, electromagnetic, or digital signals, that may be transmitted via a communication medium such as a network and / or a wireless link.

[0107] As mentioned above, it should be understood that parts of the described solution can be implemented in the vehicle 1, in a system arranged outside the vehicle 1, or in a combination of inside and outside the vehicle; for example, in a server 15 in communication with the vehicle, i.e. a so-called cloud solution. Different features and steps of the embodiments can be combined in other combinations than those described.

[0108] It should be noted that any reference signs do not limit the scope of the claims, that the present invention may be implemented at least partially by both hardware and software, and that several "means" or "units" may be represented by the same item of hardware.

[0109] Although the drawings may show a specific order of method steps, the order of the steps may be different from the order depicted. In addition, two or more steps may be performed simultaneously or partially concurrently. This variation will depend on the selected software and hardware systems and designer selections. All these variations are within the scope of the present invention. The embodiments mentioned and described above are given as examples only and should not be limited to the present invention. Other solutions, uses, objectives and functions within the scope of the present invention as claimed in the patent claims described should be obvious to those skilled in the art.

Claims

1. A method for determining an acceptable risk model to formulate a driving strategy of an automated driving system (ADS), the method comprising: obtaining a baseline risk model generated based on one or more risk-related parameters; Obtaining sensor data from a sensor system of a vehicle, wherein the vehicle is a vehicle operated by a human driver or a vehicle operated by an ADS, wherein the sensor data includes: sensor data of a perception and / or manipulation system of the vehicle, and sensor data associated with a driving behavior of the vehicle in a specific location and / or in a specific scenario; estimating one or more risk values ​​associated with the one or more risk-associated parameters based on the obtained sensor data; generating an adopted risk model by updating one or more model parameters of the baseline risk model for the specific location and / or scenario based on the estimated one or more risk values; and The generated adopted risk model is provided to determine the acceptable risk model for formulating the driving strategy of the ADS.

2. The method according to claim 1, wherein: The baseline risk model is generated based on one or more risk-associated parameters, the one or more risk-associated parameters including any one of the following: The statistical probability of various types of accidents or incidents occurring in the specific location and / or the specific scenario; a statistical probability and / or attribute of one or more external objects being present in the surroundings of the vehicle at the specific location and / or in the specific scene; as well as The driving behaviors of multiple vehicles operated by human drivers or multiple vehicles operated by ADS in the specific location and / or the specific scenario.

3. The method according to claim 1, wherein: The driving behavior of the vehicle operated by the human driver includes at least the driver's reaction time, acceleration, braking, steering, lane keeping or speed control and / or the driver's gaze direction and trajectory of visual attention and / or a safety margin selected by the driver; and The driving behavior of the ADS-operated vehicle includes at least acceleration, braking, steering, lane keeping or speed control controlled by the ADS of the vehicle.

4. The method according to any one of claims 1 to 3, wherein: The specific location includes at least a specific geographical location on a map and / or a specific type of location including an intersection, and wherein the specific scene includes at least an occlusion state in the surrounding environment of the vehicle.

5. The method according to claim 1, wherein: The method further comprises: The generated adopted risk model is provided to determine the acceptable risk model by transmitting the generated adopted risk model for the vehicle to a remote server for determining the acceptable risk model based on a plurality of adopted risk models transmitted by the plurality of human driver-operated vehicles or the plurality of ADS-operated vehicles.

6. The method according to claim 5, wherein: The vehicle is an ADS-equipped vehicle, and wherein the method further comprises: receiving from the remote server the determined acceptable risk model formed based on integrated information obtained from the plurality of human driver-operated vehicles and the plurality of ADS-operated vehicles; and An updated driving strategy for the ADS developed based on the determined acceptable risk model is received from the remote server.

7. The method according to claim 6, wherein: the integrated information including the generated adopted risk model of the plurality of human driver operated vehicles and the plurality of ADS operated vehicles transmitted to the remote server; and The determined acceptable risk model includes one or more globally updated baseline risk model parameters determined based on the integrated information that meets acceptable risk criteria.

8. The method according to claim 7, wherein: The acceptable risk criteria include a risk of the adoption of the human driver-operated vehicle indicative of a safety-improving performance of the human driver-operated vehicle. 9 . A non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors of a computing device of a vehicle, cause the computing device to perform the method of claim 1 .

10. A computer program product comprising instructions which, when executed by one or more processors of a computing device of a vehicle, cause the computing device to perform the method of claim 1.

11. A system for determining an acceptable risk model to formulate a driving strategy for an automated driving system ADS, wherein: The system includes a processing circuit configured to: obtaining a baseline risk model generated based on one or more risk-related parameters; Obtaining sensor data from a sensor system of a vehicle, wherein the vehicle is a vehicle operated by a human driver or a vehicle operated by an ADS, wherein the sensor data includes: sensor data of a perception and / or manipulation system of the vehicle, and sensor data associated with a driving behavior of the vehicle in a specific location and / or in a specific scenario; estimating one or more risk values ​​associated with the one or more risk-associated parameters based on the obtained sensor data; generating an adopted risk model by updating one or more model parameters of the baseline risk model for the specific location and / or scenario based on the estimated one or more risk values; and The generated adopted risk model is provided to determine the acceptable risk model for formulating the driving strategy of the ADS.

12. The system according to claim 11, wherein: The processing circuit is further configured to: providing the generated adopted risk model of the vehicle by transmitting the generated adopted risk model to a remote server to determine the acceptable risk model, Wherein, the remote server is adapted to determine the acceptable risk model based on a plurality of adopted risk models transmitted by the plurality of human driver operated vehicles or the plurality of ADS operated vehicles.

13. A system according to any one of claims 11 or 12, wherein: The vehicle is an ADS-equipped vehicle, and wherein the processing circuit is further configured to: receiving from the remote server the determined acceptable risk model formed based on integrated information obtained from the plurality of human driver-operated vehicles and the plurality of ADS-operated vehicles; and An updated driving strategy for the ADS developed based on the determined acceptable risk model is received from the remote server.

14. The system according to claim 13, wherein: the integrated information including the generated adopted risk model of the plurality of human driver operated vehicles and the plurality of ADS operated vehicles transmitted to the remote server; and The determined acceptable risk model includes one or more globally updated baseline risk model parameters determined based on the integrated information that meets acceptable risk criteria.

15. A vehicle comprising: one or more onboard sensors configured to monitor the vehicle's surroundings, handling and / or driving behavior; as well as The system according to claim 11.

Images