Method for improving an automated driving function

Abstract

The invention relates to a method (3) for improving an automated driving function that is designed to carry out automated vehicle guidance at least partially on the basis of map data. The method (3) makes provision to identify (31) manual overriding of the automated vehicle guidance of a vehicle (1, 2) and to modify (32) the map data and / or the automated driving function based on the identified overriding, for the purposes of carrying out the automated vehicle guidance in the future.

Description

[0001] Methods for improving an automated driving function

[0002] The present invention relates to a method for improving an automated driving function. Furthermore, the present invention relates to a computing device for carrying out such a method, as well as a computer program and a computer-readable storage medium.

[0003] Various driver assistance systems are known from the state of the art, which can increase safety and / or comfort for a driver or vehicle occupants by providing automated driving functions. Driver assistance systems can, for example, take over certain driving tasks or even control the vehicle autonomously within the framework of partially automated driving, highly automated driving, or fully automated driving.

[0004] Examples of driver assistance systems that automatically intervene in the vehicle's steering include longitudinal guidance systems, lateral guidance systems, and systems with coupled longitudinal and lateral guidance. Some common

[0005] Longitudinal guidance assistance systems include, for example, automatic speed limiting, automatic speed control (e.g., Dynamic Cruise Control - DCC, as used in BMW Group vehicles), or adaptive cruise control (Adaptive Cruise Control or Active Cruise Control - ACC).

[0006] Lateral guidance assistance systems, longitudinal guidance assistance systems, or coupled steering and lateral guidance assistance systems can, for example, implement a so-called hands-off or feet-off functionality, which relieves the driver of certain steering or...

[0007] Take over pedal operation tasks.

[0008] Automated driving functions generally allow the driver to manually override them. For example, with a steering and lane keeping assist system, the driver can intervene in the lateral control at any time by manually turning the steering wheel, for instance, to avoid an unexpectedly appearing object to which the automatic lateral control might not react. Furthermore, it may be possible, for example, for the driver to override an adaptive cruise control (ACC) function with the accelerator pedal and thus temporarily take over longitudinal control. When the driver then releases the accelerator pedal, the ACC function resumes maintaining a set speed or maintaining control over a vehicle ahead. For the purposes of this document, "manual" override of the automated driving function is not limited to override performed with the hands (e.g., on the steering wheel).Rather, "manual" oversteer generally refers to an intervention by the driver in the vehicle's control, which, depending on the type of automated driving function and the provided controls, may also include, for example, pedal operation with the feet.

[0009] Automated driving functions are generally performed at least partially based on environmental perception using the vehicle's environmental sensors. For example, the vehicle may be equipped with one or more cameras, radar and / or lidar sensors for this purpose.

[0010] In addition to or as an alternative to environmental sensor data, automated driving functions can also be executed based on map data. For example, there are longitudinal control automated driving functions that react to upcoming road events, such as traffic lights, turns, or speed limits, and proactively adjust the vehicle speed to a suitable level (possibly even bringing the vehicle to a standstill).

[0011] The applicant's patent DE 102016205 508 B4 describes a driver assistance system that detects upcoming relevant events requiring an adjustment of speed using a camera and / or map-based system and adjusts the longitudinal guidance accordingly.

[0012] Due to map errors, such automated driving functions can sometimes exhibit undesirable longitudinal and / or lateral steering behavior. For example, the vehicle may decelerate implausibly if an intersection or turn is incorrectly mapped. It can also happen that the driver assistance system incorrectly applies a red light, which is actually assigned to an adjacent lane, to its own lane and therefore unnecessarily initiates deceleration or even braking to a standstill.

[0013] The object of the present invention is to provide a method which at least partially overcomes the disadvantages of the prior art.

[0014] This task is solved by the subject matter of the independent claims.

[0015] Advantageous further developments are specified in the dependent claims. A first aspect of the invention relates to a method for improving an automated driving function, which is configured to perform automated vehicle control at least partially based on map data.

[0016] In this document, the terms "automated driving" and "automated vehicle control" refer to driving with automated longitudinal and / or lateral control. An automated driving function is a software-based function that enables such automated driving.

[0017] Automated driving can, for example, involve longer periods of driving on the motorway or time-limited driving during parking maneuvers.

[0018] The term "automated driving" encompasses automated driving at any level of automation. Examples of automation levels include assisted, partially automated, conditionally automated, highly automated, and fully automated driving (each with an increasing degree of automation). The five automation levels mentioned above correspond to SAE Levels 1 to 5 of the SAE J3016 standard (SAE - Society of Automotive Engineering) as of April 30, 2021. In assisted driving (SAE Level 1), the system performs longitudinal or lateral control in certain driving situations. In partially automated driving (SAE Level 2), the system takes over longitudinal and lateral control in certain driving situations, but the driver must continuously monitor the system, as with assisted driving.In conditionally automated driving (SAE Level 3), the system takes over longitudinal and lateral control in certain driving situations without the driver needing to constantly monitor the system; however, the driver must be able to take over control of the vehicle within a certain timeframe if requested by the system. In highly automated driving (SAE Level 4), the system takes over control of the vehicle in certain driving situations, even if the driver does not respond to a request to intervene, thus eliminating the driver as a fallback option. In fully automated driving (SAE Level 5), the system can perform all aspects of the dynamic driving task under any road and environmental conditions that a human driver can also handle.In one step of the method according to the invention, a manual override of the automated vehicle control carried out by a driver of a vehicle in which the automated driving function is executed is detected.

[0019] A further step of the procedure provides that, depending on the detected oversteering, the map data underlying the automated vehicle guidance and / or the automated driving function will be modified for the purposes of future execution of the automated vehicle guidance.

[0020] The vehicle in question may be, in particular, a motor vehicle. A motor vehicle in this sense is a land vehicle that is moved by mechanical power and is not bound to railway tracks. A motor vehicle can be, for example, a car, a motorcycle, or a tractor.

[0021] Manual override can include, for example, operating the accelerator or brake pedal to cause acceleration or deceleration of the vehicle that deviates from the automated driving system. Alternatively or additionally, manual override can include steering wheel input to override automated lateral control.

[0022] As an example scenario, for instance, in a curve that is incorrectly mapped as very tight, the automated driving function might be overridden by pressing the accelerator pedal and / or turning the steering wheel. Particularly if this occurs repeatedly (i.e., during multiple passages of the same vehicle or different vehicles through the same curve), it can be interpreted as an indicator that the curve is not actually as tight as indicated in the map data. Depending on this finding, a correction to the map data may then be necessary.

[0023] However, if repeated instances of the automated vehicle guidance system being overridden by braking may indicate, for example, that the digital map for the relevant section of road contains an excessively high speed limit or, in the case of a curve, an excessively large curve radius.

[0024] In addition to the detected manual override, environmental sensor data and / or vehicle dynamics data (such as odometry data) can also be used as a basis for modifying the map data and / or the automated driving function. For example, a correction of an incorrectly mapped speed limit sign can be made—in addition to the detected manual override(s)—also based on camera data recorded by one or more vehicles passing the location in question, which may show, for example, that a traffic sign indicates a maximum speed limit of 80 km / h instead of the 60 km / h stored in the map. In other words, the detected override(s) can be, for example,combined with environmental sensor data to provide a sufficient information basis for changing the map data and / or the automated driving function.

[0025] Manual override can be detected particularly in situations where the automated vehicle guidance system intervenes automatically in response to a route event recorded in the map data. In other words, manual override of such an automatic intervention can be detected.

[0026] For example, automated vehicle control can include automated longitudinal control. In this case, the automatic intervention in vehicle control can specifically include accelerating or decelerating the vehicle.

[0027] For example, manual override may involve using the accelerator pedal to accelerate the vehicle, thereby overriding an automatic braking action that, from the driver's point of view, would be implausible due to a track event recorded on the map (but which may not actually exist).

[0028] Alternatively or additionally, the automated intervention can relate to lateral guidance and in particular include an automatic steering intervention which is overridden by a manual steering intervention.

[0029] Depending on the detected override of the automatic intervention in vehicle guidance, which the automated driving function executed in response to a route event specified in the map data, the map data and / or the automated driving function can then be modified with respect to the route event for the purposes of future execution of automated driving guidance in the relevant section of the route. According to some embodiments, the route event may include at least one element from the following list: a curve; a roundabout; a turn; a motorway exit; a motorway entrance; an exit from a federal highway; an entrance to a federal highway; a crest of a road; a traffic light; a local speed limit (or its cancellation); a stop sign situation; a yield sign situation. The route event may accordingly be associated with a specific traffic sign, such as...in the form of a yield sign, a stop sign or a speed limit sign, which is to be expected at the location in question according to the map information.

[0030] According to further training, the procedure can be carried out on a fleet-based basis. This can mean, for example, that one or more manual overrides of the automated driving system are detected in several vehicles, and that the map data and / or the automated driving function (in the form of corresponding software) are modified depending on the override operations detected in the multiple vehicles.

[0031] For example, the map data and / or the current software version of the automated driving function can be stored and processed in an external storage device, such as a backend. This means that the data can be modified, in particular, depending on the detected override or override states (possibly in several different vehicles). For instance, information about detected manual override of the automated driving system—preferably in anonymized form—can be transmitted to the backend. The backend can be operated, for example, by the vehicle manufacturer.

[0032] The backend can include, for example, one or more computing facilities as well as one or more storage facilities.

[0033] In the backend, for example, increasingly relevant information for improving the automated driving function can be aggregated over time. This information could stem from, for instance, recorded override events across an entire vehicle fleet. Correspondingly corrected or adapted digital maps and / or software versions of the automated driving function can then be saved. The digital map and / or the software of the automated driving function can thus be continuously enriched with the latest information and corrected for errors. As a result, the automated driving function can be continuously improved during operation, for example, across an entire vehicle fleet. In particular, this allows the automated driving function to become increasingly error-free and enables the correction of map errors.

[0034] It is further within the scope of the invention that the modified map data and / or the modified automated driving function (in the form of updated software) are made available to at least one vehicle other than the vehicle(s) in which the manual override was detected. For example, the modified map data and / or the modified automated driving function can be made available to several vehicles. This can be done, for example, by means of an over-the-air update transmitted from a backend to a vehicle fleet (or a part thereof) of a vehicle manufacturer.

[0035] Alternatively or in addition to a fleet-based application, the procedure can also be carried out locally in a single vehicle, so that the automated driving function or the underlying digital map is changed solely on the basis of the manual override operations recorded in the vehicle in question and solely for a future execution of the automated driving function in that vehicle.

[0036] A second aspect of the invention is a computing device that is set up (i.e., in particular programmed) to execute a method according to the first aspect of the invention.

[0037] The computing device can also be an arrangement distributed across several spatially separate computing devices. For example, parts of the method can be located in one or more computing devices (such as control units) on board one or more vehicles and / or in an external backend. However, embodiments are also possible in which the computing device(s) are located solely in the backend or solely in one or more vehicles.

[0038] A third aspect of the invention is a computer program comprising instructions that, when executed by a computing device (such as a computing device according to the second aspect of the invention), cause it to execute a method according to the first aspect of the invention. A fourth aspect of the invention is a computer-readable storage medium comprising instructions that, when executed by a computing device (such as a computing device according to the second aspect of the invention), cause it to execute a method according to the first aspect of the invention. In other words, a computer program according to the third aspect of the invention can be stored on the computer-readable storage medium.

[0039] The invention will now be explained in more detail using an exemplary embodiment and with reference to the accompanying drawings. These drawings show:

[0040] Fig. 1 shows a schematic representation of an example situation in which a method for improving an automated driving function is carried out involving two vehicles and a backend.

[0041] Fig. 2 shows a schematic block diagram with steps of a procedure for improving an automated driving function.

[0042] Fig. 1 refers to an example scenario with two vehicles 1, 2, in which a method 3 according to the invention for improving an automated driving function is carried out by including a backend 6, which can be operated, for example, by a vehicle manufacturer of the two vehicles 1, 2.

[0043] The invention will now be explained by way of example using this example scenario, with reference also being made to process steps 31-33 according to Fig. 2, which shows a schematic flow diagram of process 3.

[0044] Figure 1 shows a first vehicle 1 equipped with a longitudinal automated driving function, such as adaptive cruise control. The automated driving function is controlled by a control unit 10 of the first vehicle 1. The automated driving function is configured to react proactively to route events 4 relevant to longitudinal guidance, such as changing speed limits, based on map data, and to adjust the speed of the vehicle 1 accordingly.

[0045] In the example shown, the start of a local speed limit of 60 km / h is recorded in the map data as a relevant road event (4) for longitudinal control. In other scenarios, a relevant road event, where there may be a discrepancy between reality and mapping, could relate, for example, to the presence or distance of a roundabout or other road characteristics.

[0046] It is assumed that the road segment shown in Fig. 1 is a section of a rural road where a speed limit of 100 km / h normally applies. Accordingly, the first vehicle 1 initially has a speed higher than 60 km / h. Due to map information about the upcoming start of a road segment with a speed limit of only 60 km / h, the automated driving function is prompted to automatically intervene in the longitudinal control, thereby decelerating the first vehicle 1 so that it is not traveling faster than 60 km / h when it reaches the road segment event 4.

[0047] It can happen that such a predictive deceleration in response to route event 4 seems implausible to the driver. In particular, the apparent speed limit of 60 km / h could be a map error. For example, the map might be outdated and the speed limit no longer exists or exists in a different form, such as a higher maximum speed of 80 km / h. In such a case, the driver would normally override the automatic intervention in the longitudinal control, namely the deceleration initiated by the automated driving function, by pressing the accelerator pedal, pressing a button (e.g., a rocker switch), or similar means.

[0048] The procedure involves detecting manual override of the automated vehicle guidance system (step 31 in Fig. 2). This detection (31) of the override can be performed, for example, by the control unit 10.

[0049] Information about the detected manual override is transmitted from the first vehicle 1 to the backend 6 via a mobile communication connection.

[0050] The backend 6 comprises a computing unit 61 and a storage unit 62, which enable the processing and storage of the recorded information.

[0051] The computing unit 61 receives the information about manual override operations recorded by the first vehicle 1. The storage unit 62 can serve as storage or intermediate storage for the received information or for information subsequently derived from it. Furthermore, the storage unit 62 also stores the current version of the digital map as the basis for the automated driving function and / or the current software version of the automated driving function.

[0052] In addition to the information about manual override operations recorded by the first vehicle 1, other (not shown) vehicles can also provide such information to backend 6. For example, information recorded in backend 6 from an entire fleet of vehicles can be aggregated.

[0053] Depending on the information acquired, the computing unit 61 modifies the map data and / or the software of the automated driving function with reference to the route event 4 for the purposes of a future execution of the automated vehicle guidance (step 32 in Fig. 2) and stores the modified map data and / or the modified software version of the automated driving function in the storage unit 62.

[0054] For example, in this way an incorrect mapping of the supposed speed limit 4 to 60 km / h can be corrected, especially if several vehicles in this section of the route send information about an override of an automatic deceleration intervention by the automated vehicle guidance to backend 6.

[0055] In addition to information about manual override operations, further information recorded in the first vehicle 1 or in other vehicles of the fleet, such as camera data, can also be used as a basis for changing the map data and / or the software of the automated driving function.

[0056] For example, it can be taken into account that, according to camera data from several vehicles, there is actually no traffic sign at the location in question indicating a maximum speed limit of 60 km / h (but instead, for example, no traffic sign at all or a traffic sign indicating a higher maximum speed limit than 60 km / h). This information, in combination with information about manual override operations (e.g., by accelerator pedal inputs), can be used by the computing device 61 to correct the map data and / or the software of the automated driving function so that, in a future execution of the automated driving function in the relevant section of the road, no deceleration or only a smaller deceleration (to the higher maximum speed limit) is applied.

[0057] The modified map data and / or the modified software of the automated driving function can be made available to both the first vehicle 1 and other vehicles in the fleet, which are represented by a second vehicle 2 in Fig. 1 (optional step 33 in Fig. 2). In particular, the modified map data and / or the updated software version of the automated driving function can be transmitted from the backend 6 to the second vehicle 2 (and other vehicles in the fleet) via a mobile network update over the air and used by a control unit 20 of the second vehicle 2 (and control units of the other vehicles) as the basis for the further execution of the automated driving function. Thus, the second vehicle 2 benefits from the information obtained by the first vehicle 1 and, when it approaches the position of the (presumed) road event 4, is not unnecessarily orThe delay is unnecessarily significant. As a result, the automated driving function can be continuously improved for the entire vehicle fleet during operation.

Claims

Patent claims 1. Method (3) for improving an automated driving function which is configured to perform automated driving at least partially on the basis of map data, wherein the method comprises detecting manual override of the automated driving function of a vehicle (1, 2) (31); and modifying the map data and / or the automated driving function for the purposes of future automated driving operation depending on the detected override (32).

2. Method (3) according to claim 1, wherein the manual override is detected in a situation (31) in which the automated vehicle guidance system automatically intervenes in the vehicle guidance of the vehicle (1, 2) in response to a route event (4) recorded in the map data, and wherein, depending on the detected override, the map data and / or the automated driving function are modified with respect to the route event (32).

3. Method (3) according to claim, wherein the route event (4) comprises at least one element from the following list: a curve; a roundabout; a turn; a motorway exit; a motorway entrance; a crest of a road; a traffic light; a local speed limit or its cancellation; a stop sign situation; a yield sign situation.

4. Method (3) according to one of the preceding claims, wherein the automated vehicle guidance comprises automated longitudinal guidance.

5. Method (3) according to claim 4, insofar as it relates back to claim 2 or 3, wherein the automatic intervention comprises accelerating and / or decelerating the vehicle (1, 2).

6. Method (3) according to one of the preceding claims, wherein the automated vehicle guidance comprises automated lateral guidance.

7. Method (3) according to claim 6, insofar as it relates back to claim 2 or 3, wherein the automatic intervention includes a steering intervention.

8. Method (3) according to one of the preceding claims, wherein one or more manual override operations of the automated vehicle guidance are detected in several vehicles (1 , 2) (31) and wherein the map data and / or the automated driving function are modified depending on the override operations detected in the several vehicles (1 , 2) (32).

9. Method (3) according to one of the preceding claims, wherein the modified map data and / or the modified automated driving function are provided to several vehicles (1 , 2) (33).

10. Computing device (10, 20, 61) configured to perform a method (3) according to any of the preceding claims.

11. Computer program comprising instructions which, when the program is executed by a computing device (10, 20, 61), cause it to execute a method (3) according to any one of claims 1 to 9.

12. Computer-readable storage medium comprising instructions which, when executed by a computing device (10, 20, 61), cause it to execute a method (3) according to any one of claims 1 to 9.

Images