Automated Driving System

Abstract

An automated driving system (ADS) includes an automation control module for controlling one or more driving functions of a vehicle, and a safety control module for determining one or more operating conditions relevant to the safety performance of the vehicle, such as driver awareness and processor temperature. The automation control module is configured to automatically adjust the speed of the vehicle based on the one or more operating conditions determined by the safety control module.

Claims

1. An automated driving system, comprising: an automation control module for controlling one or more driving functions of a vehicle; and a safety control module for determining one or more operating conditions relevant to a safety performance of the vehicle, wherein the automation control module is configured to automatically adjust a speed of the vehicle based on the one or more operating conditions determined by the safety control module.

2. The automated driving system of claim 1, wherein the automation control module is further configured to reduce the speed of the vehicle in response to the one or more operating conditions indicating a reduced safety performance scenario at a current vehicle speed.

3. The automated driving system of claim 1, wherein the one or more operating conditions relevant to the safety performance of the vehicle comprises driver awareness and processor temperature.

4. The automated driving system of claim 1, wherein the automation control module is further configured to adjust the speed of the vehicle by a pre-determined level based on the one or more operating conditions.

5. The automated driving system of claim 1, wherein the automation control module is further configured to adjust the speed of the vehicle by a dynamic level based on the one or more operating conditions.

6. An automated driving system, comprising: a driver awareness module for determining a level of driver awareness based on one or more driver awareness inputs; and an automation control module for controlling one or more driving functions of a vehicle and being operable at a lower automation level and a higher automation level, wherein the automation control module is configured to activate the higher automation level in response to the driver awareness module determining that the level of driver awareness is below an awareness threshold.

7. The automated driving system of claim 6, wherein the higher automation level is associated with a limitation of one or more control parameters.

8. The automated driving system of claim 7, wherein the limitation of one or more control parameters comprises limiting a speed of the vehicle.

9. The automated driving system of claim 6, wherein the lower automation level corresponds to level 1 or 2, and the higher automation level corresponds to level 4 automation or higher.

10. An automated driving system, comprising: an automation control module for controlling one or more driving functions of a vehicle; and a temperature control module for identifying a processor overheating scenario, wherein the processor overheating scenario leads to the automation control module exceeding an operating temperature threshold, wherein the automation control module is configured to reduce a speed of the vehicle in response to the temperature control module identifying the processor overheating scenario.

11. The automated driving system of claim 10, wherein the temperature control module is configured to identify the processor overheating scenario based on a detected processor rate exceeding a threshold.

12. The automated driving system of claim 11, wherein the temperature control module is configured to identify the processor overheating scenario based on the detected processor rate exceeding the threshold for a predetermined duration.

13. The automated driving system of claim 10, wherein the temperature control module is configured to identify the processor overheating scenario based on a detected temperature exceeding a threshold.

14. The automated driving system of claim 13, wherein the temperature control module is configured to identify the processor overheating scenario based on the detected temperature exceeding the threshold for a predetermined duration.

15. The automated driving system of claim 10, wherein the temperature control module is configured to identify a processor overheating scenario based on identifying a complex driving situation.

16. The automated driving system of claim 15, wherein the temperature control module is configured to identify the complex driving situation based on one or more of upcoming road data, current weather data, and a number of objects being tracked.

17. The automated driving system of claim 16, wherein the temperature control module is configured to identify the complex driving situation based further on a complexity metric exceeding a threshold.

18. The automated driving system of claim 17, wherein the temperature control module is configured to the complexity metric based on two or more of a road complexity rating, a camera visibility distance, a road surface rating, the number of objects being tracked, and past performance data.

19. The automated driving system of claim 10, wherein the temperature control module is further configured to identify a processor capacity scenario that leads to the automation control module having capacity to increase a processor rate while maintaining a temperature below the operating temperature threshold.

20. The automated driving system of claim 19, wherein the automation control module is configured to increase the speed of the vehicle in response to the temperature control module identifying the processor capacity scenario.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Illustrative embodiments will now be described with reference to the accompanying drawings in which:

[0034] FIG. 1 shows a schematic illustration of an automatic driving system according to an embodiment;

[0035] FIG. 2 shows a flow diagram of a driver awareness algorithm employed by the automatic driving system shown in FIG. 1;

[0036] FIG. 3 shows a flow diagram of a first processor temperature protection algorithm employed by the automatic driving system shown in FIG. 1; and

[0037] FIG. 4 shows a flow diagram of a second processor temperature protection algorithm employed by the automatic driving system shown in FIG. 1.

DETAILED DESCRIPTION

[0038] FIG. 1 shows a schematic illustration of an automatous driving system (ADS) 1 according to an embodiment. In this embodiment, the ADS 1 is implemented as part of the vehicle's Electronic Control Unit (ECU) and is for use in a car having assisted and automatous driving capabilities. As such, the ADS 1 will comprise various perception systems used to monitor the driving environment and will automatically implement driving functions based on the determined conditions. The ADS 1 is operable to implement different levels of automation and, in particular, switch between a low automation level mode associated with SAE levels one or two (advanced driver assistance systems, ADAS), and a high level automation mode associated with SAE levels three and above. In this embodiment, the ADS 1 is capable implementing level two and level four automation. When level two automation is implemented, the driver is legally required to continue to monitor the driving environment, but the ADS 1 assumes control of both steering and vehicle speed functions. When SAE level four automation is implemented, the ADS is responsible for monitoring the driving environment and controls all driving functions, albeit with certain restricted parameters. In this embodiment, the vehicle speed is restricted when operating under higher, level four, automation mode.

[0039] The ADS 1 includes a number of input modules 101-105 which process data from a plurality of sensors to make determinations about the vehicle's operating condition and the environment within which it is driving. In this embodiment, the inputs include: [0040] Driver awareness detector 101 which determines an awareness level of the driver based on, for example, sensor inputs from one or more of a driver eye tracking camera, a steering wheel contact sensor, a responsiveness monitor, and a lane tracking system; [0041] ECU processing load estimator 102 which estimates the current load being processed by one or more of the processors within the ECU implementing the ADS. This estimate is used to identify when the processing load may exceed a threshold which could then lead to the processor overheating; [0042] ECU thermal load estimator 103 which estimates the temperature of one or more of the processors within the ECU for identifying when the operating temperature has exceeded a threshold which could then lead to critical processor operations being compromised. [0043] Road situation analyser 104 for determining the current driving environment external to the vehicle based on, for example, inputs from one or more of road map data, global positioning system (GPS) data, other vehicle and an object tracking sensors, road sign and traffic light cameras, and weather condition data; and [0044] Speed limit detector 105 for determining the current vehicle speed limit within the driving environment. This may be based on, for example, inputs from one or more of road map data, GPS data, and road sign cameras.

[0045] The input modules feed various processing paths of the ADS processor 109 for processing. That is, the input modules feed algorithm modules which subsequently control parameters which influence how the ADS algorithms control driving functions.

[0046] In this embodiment, the processing paths include: [0047] ECU speed limiter 106 which determines actions in relation to limiting the processing load applied to the ADS processor 109; [0048] ADS level selector 107 which determines the level of automation implemented by the ADS processor 109; and [0049] Driving speed limit decision maker 108 which determines the vehicle speed level applied by the ADS 1.

[0050] The ADS 1 provides safety functions in relation to the operating conditions of driver awareness and processor temperature which impact the safe driving performance of the vehicle. These aspects will be described in further detail below in relation to FIGS. 2 to 4.

[0051] The driver awareness function of the ADS 1 sets the level of autonomy of the vehicle based on the awareness state of the driver. As such, when the driver is aware, the ADS 1 may provide them with as much autonomy as they wish. For instance, the driver may operate the vehicle manually or engage the ADS 1 to provide level two automation. In this state, the continued driver awareness allows for minimal restrictions on driving parameters. For example, a driver may manually bring the vehicle up to a relatively high speed of say 140 km/h, depending on the current legal speed limit, and once the ADS is engaged, this speed will be maintained. However, if the ADS 1 subsequently detects that the driver has lost awareness, then the ADS automatically switches to the higher level of driving autonomy and takes steps to manage those autonomous driving conditions by, in this embodiment, significantly reducing the speed.

[0052] To explain this further, FIG. 2 shows a flow diagram of the driver awareness algorithm employed by the ADS 1. In the example scenario, in step 201, the driver may activate driver assistance, and the ADS implements level 2 automation where the ADS 1 provides adaptive cruise control and lane keeping automation. As such, the vehicle may maintain its current speed of 140 km/h for highway conditions. The ADS 1 monitors driver awareness through the Driver Awareness Detector module 101 in step 202. If it is detected that the driver has fallen asleep or is otherwise distracted, the ADS Mode Selector 107 switches the ADS Processor 109 to the higher automation level in step 203. However, the driving parameter of vehicle speed is limited at the higher automation level and hence the vehicle is automatically slowed down. For example, based on the Speed Limit Detector 105 determining a current speed limit of 140 km/h, the safe speed level at level four automation may be set at 80 km/h, for example. At the same time, the reduction in vehicle speed is performed in collaboration with the perception systems of the ADS 1 for ensuring the speed reduction is carried out in a safe manner depending on conditions, such as the level of traffic. As part of this, other complementary actions may also be enacted. For instance, the ADS 1 may indicate and change road lanes into the slow lane for allowing other traffic to pass. The speed reduction thereby provides a silent warning to the driver, whilst improving the safety of the vehicle, at the cost of a relatively small reduction in vehicle speed. If the diver regains awareness in step 204, they may then re-initiate manual driving in step 205 and choose to bring the vehicle back up to speed before reengaging the lower level, driver assistance, automation in step 201. In other alternative embodiments, the ADS 1 may instead reengage the lower level, driver assistance, in response to the driver regaining awareness, and then automatically bring the vehicle back up to speed.

[0053] Accordingly, with embodiments, when the driver's awareness drops below a set threshold, the level of autonomous driving is increased to compensate, albeit at the expense of driving speed. As such, the ADS 1 maintains safety until the driver has fully regained awareness. Non-essential comfort related functions may also be disabled when the higher automation level is activated in order to compensate for the additional ECU processing demands this imposes. Accordingly, the vehicle's available performance is effectively restricted based on the situational awareness of the driver.

[0054] The temperature protection function of the ADS 1 acts to avoid overheating by reducing the driving speed, which, in turn, reduces the data processing load required to manage the driving conditions and environment. For example, account may be taken for the external temperature conditions in which the ECU operates, as well as other weather conditions, and the complexity of the driving environment, including the number and type of objects the ADS 1 has to track at any given moment of time. In an illustrative scenario, the ADS 1 implementing the higher automation level may be traveling at a speed of 80 km/h. If it is a particularly hot summer day and the road is busy, there may be a risk of the ECU overheating because the cooling airflow over the ECU is already warm and the processor is tasked with tracking a large number of vehicles moving at relatively high speeds. To mitigate this risk, the temperature protection function responds by reducing the vehicle speed down to, for example, 60 km/h. As such, data analysis may be performed at lower rate (e.g. a reduced frame rate), thereby reducing the heat emitted by the ECU. The step of reducing the vehicle's speed is performed in collaboration with the perception systems of the ADS 1 to ensure the speed reduction is carried out in a safe manner depending on conditions. Other complementary actions, such as changing the road lane, may also be enacted as part of this process, as necessary.

[0055] To explain this further, FIGS. 3 and 4 show flow diagrams of two processor temperature protection algorithms implemented by the ADS 1. In the first process shown in FIG. 3, the current ECU processing load is monitored using ECU Processing Load Estimator 102, along with the current ECU temperature using ECU Thermal Load Estimator 103. While automated driving is activated, if the processing load exceeds a threshold (step 302) the driving speed is automatically reduced in step 303. Equally, if the if the ECU temperature exceeds a threshold (step 305), the driving speed is automatically reduced, and non-essential processing functions may also be deactivated to quickly minimise the processing burden. As such, an overheating scenario, which would otherwise lead to the ECU overheating, may be detected by the load or the temperature exceeding a threshold, and then mitigated by reducing the vehicle speed, which in turn minimises the processing demands imposed on the ECU. This may thereby provide a low-cost safety improvement with minimal impact on perceived performance of the overall system because such overheating scenario situations would happen relatively rarely.

[0056] The process shown in FIG. 3 further monitors for instances where the current ECU processing load is under-capacity. When an under-capacity scenario is detected, either because the processing load or the ECU temperature are much lower their operating thresholds (e.g. less than 50% of their capacity), the driving speed may be automatically increased in steps 305 and 309. Non-essential processing functions may also be activated. In this way, particularly in instances where the driving speed was pre-emptively reduced to avoid overheating, the driving speed may be increased according to capacity in order to optimise performance.

[0057] The ADS 1 may also account for other driving functions. For instance, as shown in FIG. 3, step 310, if the driver or ECU has activated an economy (ECO) driving mode in order to conserve battery power in an electric vehicle, the ADS 1 may decrease the driving speed and turn off non-essential processing functions. The system thereby reduces the driving speed and/or the scope of functions performed by the ECU block in order to maximize the remaining range of the vehicle. Conversely, if ECO driving mode is off, a higher maximum speed is enabled, along with non-essential functions.

[0058] FIG. 4 shows a flow diagram of a second processor temperature protection algorithm implemented by the ADS 1. This algorithm provides pre-emptive action in the event that a complex road situation is impending. That is, if the detected road conditions and environment indicate an imminent increase of the ECU load which could otherwise lead to the ECU overheating (step 402), advance action is taken to reduce the vehicle speed (step 403), to thereby mitigate that risk by reducing the processing demands. For example, if the vehicle is approaching an area, such as a busy intersection or roundabout, in which a large number of other vehicles need to be tracked simultaneously and the road arrangement is relatively complex, the ADS 1 may identify this as a Complex Driving Situation. This may be determined based on, for example, map data together with camera analysis of the current road situation. It will be understood that the number of objects needing to be tracked, such as other vehicles, pedestrians and cycles, will typically correlate with denser areas. Therefore, the ADS 1 will typically slow the vehicle down in more urban areas. Equally, if the weather conditions indicate that driving visibility is poor and/or surface traction is reduced, normal intersections or junctions may also be identified as Complex Driving Situations. In cases where a Complex Driving Situation is identified, the vehicle's speed is reduced in response to mitigate the risk of ECU overheating.

[0059] As shown in step 404, the ECU Process Estimator Module 103 continues to monitor ECU load. In instances where, despite no Complex Driving Situation having been identified, the ECU load exceeds a threshold (e.g. more than 80% of capacity used in the last 1 second period), the Speed Limit Decision Maker 108 may act to reduce the vehicle speed. Conversely, in situations where the ECU load is well below the capacity (e.g. less than 50% of capacity used in the last 1 second period), the Speed Limit Decision Maker 108 may act to increase the vehicle speed. In embodiments, past performance data may be used to enhance the prediction of future Complex Driving Situations by adjusting the complexity threshold at which situations are identified as complex. For instance, an artificial intelligence (AI) neural network may be used to enhance this prediction by monitoring ECU load during different driving situations and learning from these historical instances.

[0060] Accordingly, with embodiments, in situations where ECU overheating is detected or predicted, the ADS 1 decreases the driving speed in a controlled manner in order to reduce the required data processing rate. For example, at lower vehicle speeds, the ECU may reduce the number of frames processed per second, or lower the processor's clock frequency. Non-essential functions, namely those that are not related to driving safety (e.g. active suspension), may also be temporarily disabled. For instance, if the CPU temperature reaches levels close to 100 degrees Celsius, where the upper limit is 110 degrees, the ADS 1 will reduce the vehicle's speed. As a result, the load of the processor decreases, so that the CPU clock can be reduced, or a number of processing cores can be switched off. At the same time selected functions not related to driving safety are turned off. As a result, the heat emission from the ECU is reduced, thereby allowing the ECU to cool. When the ECU temperature returns to a safe level, the ADS 1 returns the driving functions to full performance.

[0061] Accordingly, with embodiments, the ADS 1 may operate up to the maximum legal speed for given road segment under normal conditions, but the vehicle's speed is automatically reduced under unfavourable weather conditions, and/or if the driver loses awareness, and/or if ECU is at risk of overheating. Safety is thereby improved in situations where the driver loses awareness, and/or the ECU would otherwise be at risk of overheating. At the same time, the costs associated with cooling systems for the ECU may be minimised because the risk of ECU overheating is mitigated.

[0062] It will be understood that the embodiments illustrated above shows an application only for the purposes of illustration. In practice, embodiments may be applied to many different configurations, the detailed embodiments being straightforward for those skilled in the art to implement.

[0063] In this connection, for example, it will be understood that although the illustrative example has been described as being implemented as part of the ECU, it will be understood that it may be provided as one or more components in a system which interfaces with the ECU. Equally, embodiments may be provided as software for processing by a processor within the ECU or other multi-domain controller.