Method for assisting a malfunctioning autonomous vehicle

Abstract

A method is provided for assisting operation of a malfunctioning autonomous vehicle (2) in a work environment (1). The work environment (1) includes at least one area (4). At least one manual actor (3) operates in and / or traverses the at least one area (4). The method includes obtaining diagnostic data indicative of a malfunction of the autonomous vehicle (2). The method further includes determining a fault associated with the autonomous vehicle (2) based on the diagnostic data. The method further includes determining a preventive action among a set of preventive actions for protecting the at least one area (4) based on the determined fault. The set of preventive actions includes any one or combination of: triggering the autonomous vehicle (2) to refrain from crossing and / or operating in the at least one area (4) or increasing safety requirements for crossing and / or operating in the at least one area (4), and triggering an adaptation of the road infrastructure (30) in the at least one area (4), which adaptation prompts or triggers a restriction of the movement of at least one manual actor (3) in the at least one area (4).

Description

[Technical field]

[0001] The present invention relates to a control unit and a control method for assisting the operation of a malfunctioning autonomous vehicle. The autonomous vehicle operates in a working environment shared with at least one manual actor.

[0002] The present invention may be applied to large vehicles such as trucks, buses, and construction equipment, etc. Although the present invention is described with respect to autonomous vehicles, the present invention is also applicable to semi-autonomous vehicles. [Background technology]

[0003] When designing and implementing an Autonomous Driving System (ADS), it is necessary to ensure that the system design is safe and can handle various types of problems that may occur. One of these problems is related to how to handle individual faults that occur to the autonomous vehicle during operation. When an autonomous vehicle operates in an area shared with manual actors, e.g., cyclists, passenger vehicles, pedestrians, etc., ensuring the safety of the manual actors is paramount. Therefore, the usual response to a fault in an operating autonomous vehicle is to perform an emergency stop. For example, if the pedestrian detector of the autonomous vehicle fails, it cannot be guaranteed that the autonomous vehicle will not collide with the manual actor. Therefore, the autonomous vehicle must be stopped. Stopping the autonomous vehicle causes a loss of productivity, as the autonomous vehicle cannot perform its assigned tasks, which may further reduce the productivity of other vehicles and manual actors. Therefore, continuous efforts are required to mitigate the loss of productivity when an autonomous vehicle experiences a fault. Summary of the Invention

[0004] It is an object of the present invention to provide a method for more efficiently dealing with malfunctioning autonomous vehicles.

[0005] According to a first aspect, the above object is achieved by a method as claimed in claim 1. There is therefore provided a method for assisting the operation of a malfunctioning autonomous vehicle in a working environment, the working environment comprising at least one area and at least one manual actor operating within and / or traversing the at least one area. The method comprises: obtaining diagnostic data indicative of a malfunction of the autonomous vehicle; - determining a fault associated with the autonomous vehicle based on the diagnostic data; and deciding a preventive action based on the determined impairment; and The preventive action is determined among a set of preventive actions for protecting the at least one domain.

[0006] The set of preventive actions is - triggering the autonomous vehicle to refrain from crossing and / or operating in at least one area or to increase safety requirements for crossing and / or operating in at least one area; - triggering an adaptation of road infrastructure in at least one area, the adaptation prompting a restriction of movement of at least one manual actor in the at least one area; Any one or combination of the following may be included:

[0007] When determining preventive actions based on a particular obstacle, it is possible to dynamically determine which preventive action(s) to trigger to provide the most productive outcome for the work environment. Regardless of the option(s) determined to be triggered, the autonomous vehicle may continue to perform operations in a manner that is safe for the at least one manual actor. When triggering the autonomous vehicle to refrain from crossing and / or operating in the at least one area, the autonomous vehicle may instead perform operations elsewhere, since other areas may still be safe for the malfunctioning autonomous vehicle to operate and / or cross. When triggering an increase in the safety requirements for crossing and / or operating in the at least one area, the autonomous vehicle may continue to operate in the at least one area and / or cross the at least one area as the increased safety requirements ensure the safety of the at least one manual actor. When triggering adaptation of the road infrastructure in the at least one region, the at least one manual actor is prompted to limit its movement so that the manual actor is not exposed to danger when the malfunctioning autonomous vehicle operates within the road infrastructure and / or crosses the at least one region.

[0008] Optionally, determining a preventive action among the set of preventive actions comprises: evaluating the productivity of each of the preventive actions in the set of preventive actions; -Determining which preventive actions were rated as most productive; Further includes:

[0009] Optionally, determining the preventive action includes deciding to trigger an adaptation of road infrastructure in at least one area. Deciding to trigger an adaptation of road infrastructure in at least one area may be a default decision. Optionally, the decision is predetermined.

[0010] Optionally, triggering the adaptation of the road infrastructure comprises: - closing a restricting member, such as a fence and / or a barrier, for entering at least one area and / or for exiting at least one area; - adjusting traffic lights and / or traffic signs in at least one area; - issuing an alert signal; Contains any one or more of the following:

[0011] In other words, the at least one manual actor may be constrained by the adaptation of the road infrastructure such that the malfunctioning autonomous vehicle can perform its actions and / or traverse the at least one territory without endangering the at least one manual actor.

[0012] Optionally, the alert signal comprises: - acoustic alert signal, - Optical alert signal, and - a warning message for display on a display in at least one area, Contains any one or more of the following:

[0013] Optionally, triggering the autonomous vehicle to refrain from crossing and / or operating within the at least one region comprises: -Rerouting an autonomous vehicle; and - refraining from performing at least one operation, the at least one operation requiring traversal and / or movement in at least one area; The invention further includes any one or more of:

[0014] Optionally, triggering the autonomous vehicle to increase safety requirements for crossing and / or operating within the at least one area comprises: -reducing the maximum speed and / or acceleration permitted for the autonomous vehicle in at least one region; Further includes:

[0015] Optionally, obtaining the diagnostic data includes receiving diagnostic data from the autonomous vehicle and / or from one or more external entities observing the autonomous vehicle; Further includes:

[0016] Optionally, the one or more external entities include at least one additional autonomous vehicle.

[0017] Optionally, the diagnostic data is indicative of one or more failed sensors in a set of sensors of the autonomous vehicle.

[0018] Optionally, the diagnostic data indicates a faulty speedometer and / or a faulty pedestrian detector.

[0019] Optionally, determining the fault includes mapping the diagnostic data to one or more faults. In this manner, the fault can be determined quickly.

[0020] Optionally, mapping the diagnostic data to one or more faults is further based on a predefined analysis of how different faults of the set of sensors affect the operation of the autonomous vehicle. In other words, when a particular sensor and / or component fails, the predefined analysis may have concluded that there will be a predefined fault, e.g., a fault in pedestrian detection. The predefined analysis may further indicate a preventive action to trigger for the predefined fault.

[0021] Optionally, mapping the diagnostic data to one or more faults is further based on a predetermined analysis of how different faults of the set of sensors affect operation within and / or across at least one region.

[0022] Optionally, at least one manual actor: At least one vulnerable road user (VRU), e.g., at least one pedestrian and / or cyclist; and At least one passenger vehicle, The at least one passenger vehicle may be a car, a truck, a bus, or any other vehicle with a human inside.

[0023] Optionally, the method further comprises: - triggering the determined preventive action, Further includes:

[0024] According to a second aspect, there is provided a control unit configured to carry out the method according to the first aspect.

[0025] According to a third aspect there is provided an autonomous vehicle comprising a control unit according to the second aspect. The autonomous vehicle may for example be a truck, a car or any other suitable autonomous vehicle.

[0026] According to a fourth aspect there is provided a control station comprising a control unit according to the second aspect. The control station may be remote, for example a server in a cloud environment, or may be local, for example included in a work environment.

[0027] According to a fifth aspect, there is provided a computer program comprising program code means for performing the method according to the first aspect, when said program is run on a computer.

[0028] According to a sixth aspect there is provided a computer program medium carrying a computer program comprising program code means for performing the method according to the first aspect, when said program is run on a computer.

[0029] Further advantages and advantageous features of the present invention are disclosed in the following description and in the dependent claims.

[0030] With reference to the accompanying drawings, below follows a more detailed description of embodiments of the invention, cited as examples. [Brief description of the drawings]

[0031] [Figure 1] 1 is a schematic diagram of a work environment according to an embodiment of the present disclosure. [Diagram 2] 1 is a flow chart illustrating a method according to an embodiment herein. [Diagram 3] FIG. 1 is a schematic block diagram illustrating an example scenario according to embodiments herein. [Figure 4] 1A-B are schematic block diagrams illustrating a control unit according to an embodiment of the present specification; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0032] 1 is a schematic diagram of a work environment 1 according to an embodiment of the present disclosure. The work environment 1 may be a construction site, a warehouse, and / or any other environment in which an autonomous vehicle may operate.

[0033] In the work environment 1, an autonomous vehicle 2 operates. The autonomous vehicle 2 may be any type of autonomous or semi-autonomous vehicle, such as, for example, one of a car, a truck, a bus, or a larger vehicle such as a bulldozer, a loader, a dump truck, or a concrete vehicle.

[0034] The autonomous vehicle 2 may be configured to perform one or more operations, such as a task or mission performed at least in part by operating within the work environment 1 and / or traversing an area of ​​the work environment 1. For example, the autonomous vehicle 2 may be configured to perform one or more operations in the set of areas 50. Exemplary areas in the set of areas 50 may be located within or associated with a warehouse, a dock, a quarry, and / or a loading bay. Each area in the set of areas 50 may be associated with at least one particular operation that may be performed by the autonomous vehicle 2. The one or more operations may include any suitable operation for an autonomous or semi-autonomous vehicle. For example, the one or more operations may include loading and / or unloading an item / object and / or traversing one or more areas in the set of areas 50, for example, to transport one or more objects between areas in the set of areas 50. The one or more actions may additionally or alternatively include any action(s) associated with normal traffic operations, such as, for example, safely crossing an intersection, avoiding roadway obstacles, charging and / or refueling, parking, handover from / to a manual operation or any driving operation, etc. The one or more actions may depend on the type of autonomous vehicle 2, for example, based on the capabilities of autonomous vehicle 2.

[0035] To navigate among the set of regions 50, the autonomous vehicle 2 may be configured with the set of routes 40. The set of routes 40 and the set of regions 50 may be part of a map acquired by the autonomous vehicle 2, for example, stored locally or on some remote server.

[0036] Autonomous vehicle 2 may include a set of sensors 20. The set of sensors 20 may be required by autonomous vehicle 2 to perform one or more operations in a set of areas 50 and / or to travel on a set of routes 40. When one or more sensors in the set of sensors 20 malfunction, a fault may occur in autonomous vehicle 2. The fault may mean that one or more operations cannot be performed by autonomous vehicle 2, at least in the same way as it would be operating without the fault that occurred.

[0037] The working environment 1 includes at least one region 4 in the set of regions 50. At least one manual actor 3 is arranged to operate in and / or traverse the at least one region 4, for example, in a regular and / or periodic manner. In other words, the at least one region 4 may be a region in the working environment 1 where a human being may be present. Other regions in the set of regions 50 may be separated from the manual actor. The autonomous vehicle 2 may, for example, need to traverse and / or operate simultaneously in the at least one region 4 to perform any of one or more operations. The at least one manual actor 3 may include at least one pedestrian, at least one cyclist, at least one VRU, for example, any one or more of a car, a truck, a motorcycle, or a bus. In other words, the at least one manual actor 3 may include at least one human being in some form, e.g., driving, biking or riding a car, walking, etc., within the at least one area 4. The at least one manual actor 3 may operate in the at least one area 4 by performing any suitable operation, e.g., work related. The operations performed by the at least one manual actor 3 may include, for example, crossing roads to reach a destination, performing repairs and / or construction, and / or simply being idle in the at least one area 4.

[0038] The at least one area 4 may be arranged with a road infrastructure 30. The road infrastructure 30 may be capable of restricting the movement of the at least one manual actor 3. The road infrastructure 30 may restrict the at least one manual actor 3 from operating in the at least one area 4 and / or from crossing the at least one area 4 and / or from performing certain operations, e.g. crossing roads, in the at least one area 4. The road infrastructure 30 may be capable of restricting the movement of the at least one manual actor 3 in any suitable manner. For example, the road infrastructure 30 may comprise fences and / or barriers for restricting entry into the at least one area 4 and / or for restricting crossing roads in the at least one area 4. The road infrastructure 30 may be capable of closing the fences and / or barriers and thus restricting the movement of the at least one manual actor 3. Additionally or alternatively, the road infrastructure 30 may comprise one or more traffic lights that may be used to control where the at least one manual actor 3 can go. Furthermore, the road infrastructure may be equipped with one or more alert units which may be capable of triggering an alert signal indicating a movement restriction to at least one manual actor 3 .

[0039] The embodiments herein may be performed by a control unit 70. The control unit 70 may be included in the autonomous vehicle 2, but may also be included in any other suitable location, such as, for example, a control station 80 at a remote location. The control unit 70 may be communicatively coupled to the control station 80 and / or the autonomous vehicle 2. The control unit 70 may also be communicatively coupled to the set of sensors 20 and / or the road infrastructure 30. Additionally or alternatively, the control unit 70 may be part of or communicatively coupled to any other control unit capable of controlling the autonomous vehicle 2 and / or the road infrastructure 30.

[0040] FIG. 2 illustrates a method for assisting the operation of an autonomous vehicle 2 in a work environment 1 according to an embodiment herein. The autonomous vehicle 2 is malfunctioning, e.g., any one or more of the sensors in the set of sensors 20 may be faulty. The work environment 1 includes at least one area 4. At least one manual actor 3 operates within and / or traverses the at least one area 4. The at least one manual actor 3 may include any one or more of at least one VRU, e.g., at least one pedestrian and / or cyclist, and at least one passenger vehicle. The method includes the following actions, which are described below and may be performed in any suitable order. Optional actions are indicated by dashed boxes in FIG. 2.

[0041] Action 201 The method includes obtaining diagnostic data indicative of a malfunction of autonomous vehicle 2. The diagnostic data may be any type of data that may indicate that autonomous vehicle 2 is malfunctioning. For example, the diagnostic data may include sensor data from set of sensors 20. Additionally or alternatively, the diagnostic data may include self-diagnostic information from one or more sensors in the set of sensors.

[0042] In some embodiments, obtaining the diagnostic data further includes receiving the diagnostic data from the autonomous vehicle 2 and / or from one or more external entities observing the autonomous vehicle 2. The external entities may observe one or more issues with the autonomous vehicle 2 that are indicative of a particular type of malfunction. For example, the one or more external entities may measure and / or detect that the autonomous vehicle 2 is traveling faster than expected and / or intended within a particular route or area.

[0043] In some embodiments, the one or more external entities include at least one additional autonomous vehicle.

[0044] In these embodiments, the at least one additional autonomous vehicle and autonomous vehicle 2 may be configured to share data with each other regarding the detected dynamic object. When the at least one additional autonomous vehicle detects an unreported dynamic object that autonomous vehicle 2 recently drove past, this may indicate that autonomous vehicle 2 is malfunctioning. The at least one additional autonomous vehicle may then signal to autonomous vehicle 2 and / or control unit 70 that autonomous vehicle 2 is malfunctioning.

[0045] In other words, diagnosis of the autonomous vehicle 2 may be performed by centralized cooperative monitoring, for example by an external entity, and / or by self-diagnosis within the autonomous vehicle 2.

[0046] In some embodiments, the diagnostic data indicates one or more failed sensors in the set of sensors of the autonomous vehicle 2. For example, one or more sensors in the set of sensors 20 may not report any sensor data, thus indicating a malfunction. Additionally or alternatively, two sensors in the set of sensors 20 may be used for redundancy, where the sensors report different data. Thus, at least one of them is likely corrupted. Additionally or alternatively, the diagnostic data may indicate a malfunction by indicating that at least one sensor in the set of sensors 20 switches, for example, between an object detected state and an object not detected state faster than an acceptable / allowable switching boundary.

[0047] In some embodiments, the diagnostic data is indicative of a faulty speedometer and / or a faulty pedestrian detector. These are two examples, but many other diagnostic data indicative of a malfunction may exist. For example, the diagnostic data may relate to a failure of a perception system of the autonomous vehicle 2, but may also relate to a failure of a motion control component, such as, for example, detection of a suspension defect.

[0048] Action 202 The method includes determining a fault associated with the autonomous vehicle 2 based on the diagnostic data.

[0049] Based on the determined obstacles, it may be possible to determine that safety requirements need to be enhanced for the autonomous vehicle 2 to safely operate in and / or traverse the at least one area 4, e.g., the maximum allowed speed of the autonomous vehicle 2 needs to be reduced. It may also be possible to determine whether separation of the autonomous vehicle 2 from the at least one manual actor 3 is required by other means.

[0050] Determining the obstacles may include determining whether autonomous vehicle 2 can be operated, i.e., whether autonomous vehicle 2 has essential driving capability. Determining whether autonomous vehicle 2 can be operated may further include determining whether it is safe to operate at least within and / or traverse area 4 using normal safety requirements.

[0051] The determined obstacle may further be associated with a set of capabilities of the autonomous vehicle 2. The set of capabilities may indicate which operations the autonomous vehicle 2 is capable of when operating with the determined obstacle. The set of capabilities may indicate, for example, any one or more of a maximum allowable speed and / or acceleration of the autonomous vehicle 2, one or more available sensors for use by the autonomous vehicle 2, one or more available systems for use by the autonomous vehicle 2, such as a braking system and / or a pedestrian detection system, available traffic operations for the autonomous vehicle 2 to perform, such as a left turn and / or a right turn, and / or other functions of the autonomous vehicle 2. The set of capabilities may also indicate operations the autonomous vehicle cannot perform when operating with the determined obstacle. In some embodiments, the set of capabilities may indicate, for example, a maximum speed at which the autonomous vehicle 2 is permitted to drive and may detect a pedestrian within a predetermined period of time. Another example is that a determined obstacle in a particular sensor of the set of sensors 20 may affect the tracking performance of the autonomous vehicle 2 when following a route in the set of routes 40. Thus, the set of capabilities may indicate the tracking performance of the autonomous vehicle 2. When tracking performance is poor, the autonomous vehicle 2 may require a corresponding margin for the width of roads / lanes that the autonomous vehicle 2 can safely operate on when, for example, veering off a path based on tracking performance. The set of capabilities may be determined by mapping fault and / or diagnostic data to a lookup table or database.

[0052] In some embodiments, determining the fault includes mapping the diagnostic data to one or more faults. In this manner, faults can be quickly mapped from the diagnostic data. In some embodiments, mapping the diagnostic data to one or more faults is further based on a predefined analysis of how different faults of the set of sensors affect operation of the autonomous vehicle 2. For example, if the diagnostic data indicates that a combination of sensors has failed, the predefined analysis may indicate which fault is associated with the diagnostic data. In some embodiments, mapping the diagnostic data to one or more faults is further based on a predefined analysis of how different faults of the set of sensors affect operation within at least one region and / or when traversing at least one region 4.

[0053] Action 203 The method includes determining a preventive action among a set of preventive actions for protecting the at least one area 4 based on the determined failure. The determination may further be based on a set of capabilities related to the determined failure. The set of preventive actions includes the following actions:

[0054] - triggering the autonomous vehicle 2 to refrain from crossing and / or operating in at least one area 4 or to increase safety requirements for crossing and / or operating in at least one area 4; - triggering an adaptation of the road infrastructure 30 in at least one area 4, the adaptation prompting a restriction of the movement of at least one manual actor 3 in the at least one area 4.

[0055] - their combinations, i.e. a combination of both above options, are also possible.

[0056] In other words, determining the preventive action includes considering the above actions to ensure that it is possible to operate autonomous vehicle 2 in a safe manner even when autonomous vehicle 2 is malfunctioning, thereby increasing productivity and maintaining safety.

[0057] The determination of which preventive action to trigger is based on the fault, such as the type of fault. The determination may include mapping the determined fault to a preventive action, for example, by a database or lookup table.

[0058] In some embodiments, determining the preventive action includes determining to trigger an adaptation of the road infrastructure 30 in the at least one region 4. For example, the adaptation of the road infrastructure 30 in the at least one region 4 may be the default preventive action.

[0059] In these embodiments, determining an action may include triggering an immediate adaptation of the road infrastructure 30 .

[0060] In some embodiments, triggering the adaptation of the road infrastructure 30 comprises: closing restriction elements, such as fences and / or barriers, for entering and / or exiting the at least one area 4, so that the at least one manual actor 3 cannot enter the at least one area 4 simultaneously with the autonomous vehicle 2; adjusting traffic lights and / or traffic signs of at least one area 4, for example in such a way that at least one manual actor 3 is restricted from crossing a road when the autonomous vehicle 2 operates in and / or crosses the at least one area 4, issuing an alert signal, for example to alert at least one manual actor 3 that the autonomous vehicle 2 is operating in and / or crossing at least one area 4; Contains any one or more of the following:

[0061] In some embodiments the alert signal comprises: an acoustic alert signal, e.g. a warning sound; - Optical alert signals, e.g. warning lights; a display warning message on a display in at least one area 4, for example displayed on a sign and / or transmitted to at least one wireless device of a manual actor 3, such as a smartphone or a wearable device; Contains any one or more of the following:

[0062] In some embodiments, triggering the autonomous vehicle 2 to refrain from crossing and / or operating within at least one region 4 includes: - rerouting the autonomous vehicle 2; and refraining from performing at least one operation, which requires a crossing and / or an operation in at least one area 4; The invention further includes any one or more of:

[0063] For example, it may be possible not to perform the at least one action at all and instead perform another action. The at least one action may then be postponed until autonomous vehicle 2 is repaired. In some embodiments, it may be possible to perform an action using another route in set of routes 40, i.e., without traversing at least one region 4.

[0064] In some embodiments, triggering the autonomous vehicle 2 to increase safety requirements for crossing and / or operating within at least one area 4 comprises: reducing the maximum speed and / or acceleration permitted for the autonomous vehicle 2 in at least one region 4; Further includes:

[0065] For example, the problem may be related to a failure of the pedestrian detection system such that the pedestrian detection system can only safely detect pedestrians at slow speeds, creating danger when the autonomous vehicle 2 drives at an excessively high speed in at least one region 4.

[0066] In some embodiments, determining a preventive action among the set of preventive actions comprises: evaluating the productivity of each of the preventive actions in the set of preventive actions; -Determining which preventive actions were rated as most productive; Further includes:

[0067] In some embodiments, determining a preventive action may be further based on a set of associated capabilities of the determined disorder.

[0068] Estimating the productivity for each of the preventive actions may be based on the determined impairment and may depend on the work environment 1 and / or at least one area 4.

[0069] In some example scenarios, the at least one manual actor 3 is a pedestrian who wants to cross a road in the at least one region 4 because it is the quickest route, but in fact there is a tunnel a short distance away that is also an acceptable route. In these example scenarios, it may be more productive to operate the autonomous vehicle 2 in the at least one region 4 and adapt the road infrastructure 30 so that the pedestrian has to go through the tunnel instead.

[0070] However, if there are no alternative routes available for the pedestrian and the pedestrian needs to head to a location where they should, for example, be assisted with loading / unloading, it may be more productive to enhance the safety requirements of the autonomous vehicle 2 and / or change the route of the autonomous vehicle.

[0071] Additionally or alternatively, estimating the productivity for each of the preventive actions may depend on the type of autonomous vehicle 2 and / or the type of road infrastructure 30 in the at least one region 4. For example, depending on the type of road, restricting movement in the at least one region 4 by infrastructure 30 may be slow, i.e., it may take time to close a fence or barrier, and therefore may be less productive than rerouting the autonomous vehicle 2.

[0072] Action 204 The method may further comprise triggering the determined preventive action. Triggering the preventive action may be performed in any suitable manner.

[0073] The method will now be further described and illustrated in the following embodiments, which may be combined with any suitable embodiment above.

[0074] FIG. 3 illustrates an exemplary scenario of an embodiment herein. The exemplary scenario may be executed by the control unit 70 in the control station 80 and / or in the autonomous vehicle 2. In the exemplary scenario, the autonomous vehicle 2 performs a set of actions by traveling among a set of areas 50 including at least one area 4 by using a set of routes 40. Each of the set of routes 40 is associated with a respective action in the set of actions, also referred to as a mission. The set of actions includes a first action 301, a second action 302, a third action 303, and a fourth action 304. The fourth action 304 is associated with a route that traverses at least one area 4. The control unit 70 obtains diagnostic data, such as action 201. The diagnostic data may indicate that the camera-based pedestrian detection system of the autonomous vehicle 2 is faulty. Only at least one area 4 may have manual actors present, and therefore the autonomous vehicle 2 may operate in the remainder of the set of areas 50 without using the pedestrian detection system. Thus, the autonomous vehicle 2 can proceed with the actions in the set of actions that can be performed in isolation from the at least one manual actor 3. Thus, for the first, second and third actions 301, 302, 303, the autonomous vehicle 2 can perform the actions as long as the autonomous vehicle 2 does not cross at least one area 4. This may require the autonomous vehicle 2 to be rerouted, e.g. the autonomous vehicle 2 has to travel a longer distance to perform the action. To perform the fourth action 304, the autonomous vehicle 2 may need to operate within at least one area 4 and / or cross at least one area 4. Since the pedestrian detection system is faulty, the autonomous vehicle 2 cannot operate within at least one area 4 and / or cross at least one area 4 by using standard safety requirements. Thus, to perform the fourth action 304, there are two main options to consider, e.g. when determining the preventive action in action 203 above:

[0075] 1) executing a fourth action 304, for example by triggering an adaptation of the road infrastructure 30 to restrict the movement of the at least one manual actor 3, for example by closing a gate to the at least one area 4; 2) Triggering an increase in the safety requirements of the autonomous vehicle, for example if it may be possible to drive in at least one region 4 in some circumstances where the pedestrian detection system is impaired. For example, the pedestrian detection system may be impaired such that pedestrians can only be detected very close to the autonomous vehicle 2. These embodiments may allow estimating a maximum safe speed for the autonomous vehicle 2 that still allows the autonomous vehicle 2 to detect pedestrians in time, and further reducing the maximum permitted speed allowed by the autonomous vehicle 2 below the estimated maximum safe speed.

[0076] Preventive Actions In some embodiments herein, the set of preventative actions, for example in actions 203, may further include triggering any one or more of the following: At least one manual actor 3 is not allowed to cross a road in at least one region 4 while the autonomous vehicle 2 is within a certain number of meters. This can be achieved, for example, by restricting the movement of the at least one manual actor, as in action 203.

[0077] At least one manual actor 3 is not allowed to leave the vehicle or is restricted to using vehicles that meet certain safety levels and / or safety requirements.

[0078] The at least one manual actor 3 is limited to a predefined speed and / or a predefined traffic behavior, e.g. no left turns are allowed. This may ensure that the autonomous vehicle 2 and the at least one manual actor cannot collide, since the autonomous vehicle 2 may also be simultaneously configured with a complementary traffic behavior that causes this effect.

[0079] The at least one manual actor 3 cannot approach the autonomous vehicle 2 while it is powered on. This can be achieved, for example, by optical signaling on the autonomous vehicle 2 or on the road infrastructure 30. Another example can be communication to a wireless device of the at least one manual actor 3. In some embodiments, the current restrictions are displayed on a display of the road infrastructure 30 at the entrance to the at least one area 4, e.g. messages like "No entry", "Keep distance" or "Stay off the road".

[0080] Each preventive action allows work environment 1 to continue operating with some degree of productivity for the remainder of the day / shift, so that necessary problem resolution and maintenance may be deferred until operation of autonomous vehicle 2 is completed. In some embodiments, the preventive action may be determined based on the current capabilities of autonomous vehicle 2. The current capabilities may be determined based on determined impairments.

[0081] Offline analysis and online Some actions of the embodiments herein may be performed while the autonomous vehicle 2 is performing one or more operations in the work environment 1. This may be referred to as the online portion of the embodiments herein. The online portion may include the following steps:

[0082] - diagnostics of the sensor(s), e.g. actions 201 and 202 above; -Adapting to limited capabilities, for example actions 203 and 204 above.

[0083] There may be different options for preventive actions, for example those that affect the operation of the autonomous vehicle 2 and those that adapt to the road infrastructure 30 of at least one area 4.

[0084] Some actions of the embodiments herein, particularly with respect to analysis of failures and their capabilities, may be performed at a remote location and / or at any suitable early point before performing one or more operations. These actions may be referred to as the offline portion of the embodiments herein. The offline portion may evaluate the capabilities of the autonomous vehicle 2 given a particular sensor failure and how they may affect possible operations, missions, and / or the working environment 1. The offline portion may include evaluating how the particular sensor failure affects the capabilities of the working environment 1 and / or the operating domain of the autonomous vehicle 2. The effect is analyzed for how it affects the operation of the autonomous vehicle 2 and how it affects at least one area 4 with respect to the autonomous vehicle 2 and / or at least one manual actor 3.

[0085] To perform the embodiments described herein, the control unit 70 may be configured to perform any one or more of the actions 201-204 above, or any of the other examples or embodiments described herein. The control unit 70 may include, for example, the configuration shown in Figures 4a and 4b.

[0086] The control unit 70 may include an input / output interface 400 configured to communicate with any necessary components or entities of the embodiments herein. The input / output interface 400 may include a wireless and / or wired receiver (not shown) and a wireless and / or wired transmitter (not shown). The control unit 70 may be located in any suitable location on the autonomous vehicle 2 and / or in the control station 80.

[0087] The control unit 70 may further be configured to acquire, for example by an acquisition unit 401 in the control unit 70, diagnostic data indicative of a malfunction of the autonomous vehicle 2, for example as in action 201 above.

[0088] The control unit 70 may further be configured to determine, for example by a determination unit 402 in the control unit 70, a fault associated with the autonomous vehicle 2 based on the diagnostic data, for example as in action 202 above.

[0089] The control unit 70 may further be configured to determine a preventive action among a set of preventive actions for protecting at least one area 4, such as action 203 above, based on the determined fault, for example by a decision unit 403 in the control unit 70.

[0090] The control unit 70 may further be configured to trigger the determined preventive action, such as, for example, action 204 above, for example by a trigger unit 404 in the control unit 70 .

[0091] The embodiments herein may be implemented via a processor, such as processor 460 of the processing circuitry in the control unit 70 shown in FIG. 4a, with computer program code for performing the functions and actions of the embodiments herein. The above-mentioned program code may also be provided as a computer program medium, for example in the form of a data computer readable medium that carries computer program code for performing the embodiments described herein when loaded into the control unit 70. One such computer readable medium may be in the form of a memory stick. Furthermore, the computer program code may be provided on a server as pure program code and downloaded to the control unit 70.

[0092] The control unit 70 may further comprise a memory 470 comprising one or more memory units. The memory 470 includes instructions executable by a processor in the control unit 70. The memory 470 is arranged to be used to store, for example, information, instructions, data, configurations, trajectories, positions, regions of interest, sensor data, and applications for performing the methods herein when executed in the control unit 70.

[0093] In some embodiments, the computer program 480 comprises instructions that, when executed by a computer, such as the at least one processor 460, cause the at least one processor of the control unit 70 to perform the actions 201-204 described above.

[0094] In some embodiments, the computer readable storage medium 490 includes the respective computer program 480. The computer readable storage medium 490 may include program code for performing steps of any one of the above-mentioned actions 201-204 when said program product is executed on a computer, for example on at least one processor 460.

[0095] Those skilled in the art will appreciate that the units within control unit 70 described above may refer to one or more processors configured with a combination of analog and digital circuitry and / or software and / or firmware executed by a respective one or more processors, such as the processors described above, e.g., stored in control unit 70. One or more of these processors and other digital hardware may be included in a single application specific integrated circuit (ASIC), or several processors and various digital hardware may be distributed across several separate components, whether packaged separately or assembled into a system on a chip (SoC).

[0096] Naturally, the invention is not limited to the embodiments described above and illustrated in the drawings, but rather may be subject to many variations and modifications within the scope of the appended claims, as will be apparent to those skilled in the art.

Claims

1. A method for assisting operation of a malfunctioning autonomous vehicle (2) in a working environment (1), the working environment (1) including at least one area (4), at least one manual actor (3) operating in and / or traversing the at least one area (4), the method comprising: - obtaining (201) diagnostic data indicative of a malfunction of said autonomous vehicle (2); - determining (202) a fault associated with the autonomous vehicle (2) based on the diagnostic data; - determining (203) a preventive action among a set of preventive actions for protecting said at least one area (4) based on said determined failure, said set of preventive actions comprising: - triggering the autonomous vehicle (2) to refrain from crossing and / or operating in the at least one area (4) or to increase safety requirements for crossing and / or operating in the at least one area (4); - triggering an adaptation of the road infrastructure (30) in said at least one area (4), which adaptation prompts a restriction of the movement of said at least one manual actor (3) in said at least one area (4), or - combinations thereof, said determining (203) including: The method comprising:

2. Determining the preventive action among the set of preventive actions (203), - evaluating the productivity of each of said preventive actions in said set of preventive actions; - determining the preventive action that is rated the most productive; The method of claim 1 further comprising:

3. 10. The method according to any one of the preceding claims, wherein determining (203) the preventive action comprises deciding to trigger the adaptation of the road infrastructure (30) in the at least one area (4).

4. Triggering the adaptation of the road infrastructure (30), - closing a restricting member such as a fence and / or a barrier for entering and / or exiting said at least one area (4); - adjusting the traffic lights and / or traffic signs of said at least one area (4); - issuing an alert signal; 13. A method according to any one of the preceding claims, comprising any one or more of:

5. The alert signal: an acoustic alert signal; - an optical alert signal; - a warning message for display on a display in said at least one area (4); The method of claim 4, comprising any one or more of:

6. Triggering the autonomous vehicle (2) to refrain from crossing and / or operating in the at least one area (4), - rerouting the autonomous vehicle (2); - refraining from performing at least one operation, said at least one operation requiring said traversal and / or movement in said at least one area (4); 13. The method of any one of the preceding claims, further comprising any one or more of:

7. triggering the autonomous vehicle (2) to increase safety requirements for crossing and / or operating within the at least one area (4); - Reducing the maximum speed and / or acceleration allowed for the autonomous vehicle (2) in said at least one region (4).

2. The method of any one of the preceding claims, further comprising:

8. 13. The method of any one of the preceding claims, wherein obtaining (201) the diagnostic data further comprises receiving the diagnostic data from the autonomous vehicle (2) and / or from one or more external entities observing the autonomous vehicle (2).

9. The method of claim 8 , wherein the one or more external entities include at least one additional autonomous vehicle (2).

10. 10. The method of any one of the preceding claims, wherein the diagnostic data is indicative of one or more failed sensors of a set of sensors of the autonomous vehicle (2).

11. 10. A method according to any one of the preceding claims, wherein the diagnostic data is indicative of a faulty speedometer and / or a faulty pedestrian detector.

12. 2. The method of claim 1, wherein determining (202) the fault comprises mapping the diagnostic data to one or more faults.

13. 13. The method of claim 12, wherein mapping the diagnostic data to one or more faults is further based on a predetermined analysis of how different faults of the set of sensors affect operation of the autonomous vehicle (2).

14. The method according to any one of claims 12 to 13, wherein mapping the diagnostic data to one or more faults is further based on a predetermined analysis of how different faults of the set of sensors affect operation within and / or when traversing the at least one area (4).

15. The at least one manual actor (3), at least one vulnerable road user (VRU), such as at least one pedestrian and / or cyclist; At least one passenger vehicle; 13. A method according to any one of the preceding claims, comprising any one or more of:

16. The method further comprising: - triggering said determined preventive action (204); 2. The method of any one of the preceding claims, further comprising:

17. A control unit (70) configured to carry out the method according to any one of the preceding claims.

18. An autonomous vehicle (2) comprising a control unit (70) according to claim 17.

19. A control station (80) comprising a control unit (70) according to claim 17.

20. A computer program (480) comprising program code means for carrying out the steps of any one of claims 1 to 16, when the program is run on a computer.

21. A computer program medium (490) carrying a computer program comprising program code means for performing the steps of any one of claims 1 to 16, when the program is run on a computer.

Images