Anomaly detection system and method

Abstract

A system and method for detecting anomalies of sensors of a vehicle is provided. In one embodiment, a method includes: storing a plurality of sensor correlation groups based on vehicle dynamics; processing a subset of signals based on the sensor correlation group to determine when an anomaly exists; processing a subset of signals based on the sensor correlation group to determine which sensor in the sensor correlation group is abnormal; and generating notification data based on the sensor of the abnormal correlation group.

Description

Technical Field

[0001] The present disclosure relates generally to vehicles and, more particularly, to systems and methods for detecting anomalies within a vehicle. Background Art

[0002] An autonomous or semi-autonomous vehicle is a vehicle that is able to sense its environment and navigate with little or no user input. Autonomous vehicles use sensing devices such as inertial measurement units, radar, lidar, image sensors, etc. to sense their environment. Autonomous vehicle systems also use information from global positioning system (GPS) technology, navigation systems, vehicle-to-vehicle communications, vehicle-to-infrastructure technology, and / or drive-by-wire systems to navigate the vehicle.

[0003] In addition, autonomous or semi-autonomous vehicles rely on sensor data from sensors that monitor observable states of the vehicle, such as, but not limited to, speed sensors, position sensors, yaw rate sensors, inertial sensors, etc. In some cases, one or more sensors may fail. Failures in the sensors and / or failures or malicious attacks on the corresponding sensor data can cause the control software to perceive the environment and / or vehicle functions incorrectly, resulting in the system making incorrect decisions. Therefore, it is desirable to provide improved methods and systems for detecting vehicle sensor anomalies. It is also desirable to provide methods and systems for detecting other vehicle sensor anomalies. In addition, in conjunction with the accompanying drawings and the aforementioned technical field and background technology, other desired features and characteristics of the present invention will become apparent from the subsequent detailed description and the appended claims. Summary of the invention

[0004] A system and method for detecting anomalies of sensors of a vehicle is provided. In one embodiment, a method includes: storing a plurality of sensor correlation groups based on vehicle dynamics; processing a subset of signals based on the sensor correlation group to determine when an anomaly exists; processing a subset of signals based on the sensor correlation group to determine which sensor in the sensor correlation group is abnormal; and generating notification data based on the abnormal sensor in the sensor correlation group.

[0005] In various embodiments, the plurality of sensor correlation groups include a position group, a velocity group, an acceleration group, a heading group, and a yaw rate group. In various embodiments, each of the plurality of sensor correlation groups includes one or more sensors that are correlated to the group based on vehicle dynamics. In various embodiments, the signal subset includes a position signal, a velocity signal, an acceleration signal, a heading signal, and a yaw rate signal.

[0006] In various embodiments, processing a subset of the signals to determine when an anomaly exists includes calculating a mean absolute error (MAE) based on a plurality of correlation groups.

[0007] In various embodiments, processing a subset of signals to determine when an anomaly exists is based on a defined batch size.

[0008] In various embodiments, processing a subset of signals to determine which sensor in a sensor correlation group is anomalous is based on a depth-first search approach and a plurality of sensor correlation groups.

[0009] In various embodiments, processing the subset of signals to determine which sensor in the sensor correlation group is abnormal is based on a calculated average of the predicted condition vectors associated with the sensor correlation group in the plurality of correlation groups.

[0010] In various embodiments, the method includes training a plurality of thresholds based on the abnormal signal data, and wherein processing the subset of signals to determine which sensor of the related group of sensors is abnormal is based on the plurality of trained thresholds.

[0011] In various embodiments, the method includes training a plurality of thresholds based on the abnormal signal data, and wherein processing the subset of signals to determine when a sensor of the related set of sensors is abnormal is based on the plurality of trained thresholds.

[0012] In another embodiment, a computer-implemented system is provided. The computer-implemented system includes an anomaly detection module, the anomaly detection module including one or more processors configured by programming instructions encoded in a non-transitory computer-readable medium, the anomaly detection module configured to: store a plurality of sensor correlation groups based on vehicle dynamics; process a subset of signals based on the sensor correlation groups to determine when an anomaly exists; process the subset of signals based on the sensor correlation groups to determine which sensor in the sensor correlation group is abnormal; and generate notification data based on the abnormal sensor in the sensor correlation group.

[0013] In various embodiments, the plurality of sensor correlation groups include a position group, a velocity group, an acceleration group, a heading group, and a yaw rate group. In various embodiments, each of the plurality of sensor correlation groups includes one or more sensors that are correlated to the group based on vehicle dynamics.

[0014] In various embodiments, the subset of signals includes a position signal, a velocity signal, an acceleration signal, a heading signal, and a yaw rate signal.

[0015] In various embodiments, the anomaly detection module is configured to process a subset of the signal to determine when an anomaly exists by calculating a mean absolute error based on a plurality of correlation groups.

[0016] In various embodiments, the anomaly detection module is configured to process a subset of the signals to determine when an anomaly exists based on a defined batch size.

[0017] In various embodiments, the anomaly detection module is configured to process the subset of signals to determine which sensor in the sensor correlation group is anomaly based on a depth-first search method and the plurality of sensor correlation groups.

[0018] In various embodiments, the anomaly detection module is configured to process the subset of signals to determine which sensor in the sensor correlation group is anomaly based on a calculated average of the predicted condition vectors associated with the sensor correlation group of the plurality of correlation groups.

[0019] In various embodiments, the anomaly detection module is configured to train a plurality of thresholds based on the abnormal signal data, and process a subset of signals based on the plurality of trained thresholds to determine which sensor in the related group of sensors is abnormal.

[0020] In various embodiments, the anomaly detection module is configured to train a plurality of thresholds based on the anomalous signal data, and to process a subset of signals based on the plurality of trained thresholds to determine when a sensor of the related group of sensors is anomalous. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Exemplary embodiments will be described below with reference to the following drawings, wherein like reference numerals represent like elements, and wherein:

[0022] Figure 1 is a functional block diagram illustrating a vehicle having an abnormality detection system according to various embodiments;

[0023] Figure 2 is a functional block diagram illustrating an operating environment of a vehicle having an abnormality detection system according to various embodiments;

[0024] Figure 3 is a data flow diagram illustrating an anomaly detection module according to various embodiments; and

[0025] Figure 4 and Figure 5 is a flow chart illustrating an anomaly detection method that may be performed by an anomaly detection system according to various embodiments. DETAILED DESCRIPTION

[0026] The following detailed description is merely exemplary in nature and is not intended to limit application and use. In addition, it is not intended to be bound by any express or implied theory presented in the previous technical field, background, brief overview or detailed description below. As used herein, the term module refers to any hardware, software, firmware, electronic control component, processing logic and / or processor device, individually or in any combination, including but not limited to: application-specific integrated circuits (ASICs), electronic circuits, processors (shared, dedicated or grouped) and memories that execute one or more software or firmware programs, combinational logic circuits and / or other suitable components that provide the functions.

[0027] Embodiments of the present disclosure may be described herein in terms of functions and / or logic block components and various processing steps. It should be understood that such block components may be implemented by any number of hardware, software, and / or firmware components configured to perform a specified function. For example, embodiments of the present disclosure may employ various integrated circuit components, such as storage elements, digital signal processing elements, logic elements, lookup tables, etc., which may perform various functions under the control of one or more microprocessors or other control devices. In addition, those skilled in the art will appreciate that embodiments of the present disclosure may be implemented in conjunction with any number of systems, and the systems described herein are merely exemplary embodiments of the present disclosure.

[0028] For the sake of brevity, conventional techniques related to signal processing, data transmission, signaling, control, and other functional aspects of the system (and the various operating components of the system) may not be described in detail herein. In addition, the connecting lines shown in the various figures included herein are intended to represent example functional relationships and / or physical couplings between various elements. It should be noted that many alternative or additional functional relationships or physical connections may exist in the embodiments of the present disclosure.

[0029] Reference Figure 1 According to various embodiments, an anomaly detection system, generally shown at 100, is associated with a vehicle 10. Generally speaking, the anomaly detection system 100 receives and processes signal data to detect data-centric anomalies in the vehicle 10 (ego vehicle) and other vehicles (communicating with the ego vehicle) in real time using a subset of signals from vehicle-to-everything (V2X) messages, such as, but not limited to, basic emergency and safety messages, which will be discussed in more detail below. In various embodiments, the anomaly detection system 100 traces the origin of the detected anomaly to a specific signal or group of signals. In various embodiments, the anomaly detection system 100 verifies the correctness of its own signal data and protects the vehicle 10 from anomalous / inappropriate data from other vehicles.

[0030] like Figure 1As shown in the example of FIG. 1 , the vehicle 10 is an automobile and generally includes a chassis 12, a body 14, front wheels 16, and rear wheels 18. The body 14 is disposed on the chassis 12 and substantially encloses the components of the vehicle 10. The body 14 and the chassis 12 may together form a frame. Each of the wheels 16-18 is rotatably connected to the chassis 12 near a corresponding corner of the body 14.

[0031] As shown, vehicle 10 generally includes a propulsion system 20, a transmission system 22, a steering system 24, a braking system 26, a sensor system 28, an actuator system 30, at least one data storage device 32, at least one controller 34, and a communication system 36. In various embodiments, propulsion system 20 includes an internal combustion engine, an electric machine, such as a traction motor powered solely by one or more batteries (e.g., as a pure electric vehicle) or a combination of internal combustion engines, and / or a fuel cell propulsion system (e.g., as a hybrid electric vehicle).

[0032] The transmission system 22 is configured to transfer power from the propulsion system 20 to the wheels 16-18 according to a selectable speed ratio. According to various embodiments, the transmission system 22 may include a stepped automatic transmission, a continuously variable transmission, or other suitable transmission. The braking system 26 is configured to provide braking torque to the wheels 16-18. In various embodiments, the braking system 26 may include friction brakes, wire control brakes, regenerative braking systems, such as motors, and / or other suitable braking systems. The steering system 24 affects the position of the wheels 16-18.

[0033] The sensor system 28 includes one or more sensing devices 40a-40n that sense observable conditions of the external environment and / or the internal environment of the vehicle 10. The sensing devices 40a-40n may include, but are not limited to, radar, lidar, global positioning system, optical camera, thermal camera, ultrasonic sensor, inertial measurement unit, and / or other sensors. In various embodiments, the sensor system 28 also includes one or more sensing devices 41a-41n that sense observable conditions of one or more vehicle components.

[0034] The actuator system 30 includes one or more actuator devices 42a-42n that control one or more vehicle features, such as, but not limited to, the propulsion system 20, the transmission system 22, the steering system 24, and the braking system 26. In various embodiments, the vehicle features may further include interior and / or exterior vehicle features, such as, but not limited to, doors, trunks, and cabin features, such as air, music, lighting, etc. (not numbered).

[0035] The communication system 36 is configured to wirelessly communicate information with other entities 48, such as, but not limited to, other vehicles (“V2V” communications), infrastructure (“V2I” communications), everything (“V2X” communications), remote systems, charging stations, and / or personal devices (see Figure 2 ). In an exemplary embodiment, the communication system 36 is a wireless communication system configured to communicate via a wireless local area network (WLAN) using the IEEE 802.11 standard or by using cellular data communications. However, additional or alternative communication methods, such as dedicated short-range communication (DSRC) channels or cellular V2X (C-V2X), are also considered within the scope of the present disclosure. A DSRC channel refers to a one-way or two-way short-range to medium-range wireless communication channel designed specifically for automotive use and a corresponding set of protocols and standards.

[0036] The data storage device 32 stores data for controlling the automated vehicle 10. In various embodiments, the data storage device 32 stores a defined map of the navigable environment. In various embodiments, the defined map may be predefined and obtained by a remote system (see Figure 2 34 ). For example, the defined maps may be assembled by a remote system and transmitted to the autonomous vehicle 10 (wirelessly and / or wiredly) and stored in the data storage device 32. Route information may also be stored in the data storage device 32, i.e., a set of road segments (geographically associated with one or more defined maps) that together define a route for a user to travel from a starting location (e.g., the user's current location) to a target location. It will be appreciated that the data storage device 32 may be part of the controller 34, separate from the controller 34, or part of the controller 34 and part of a separate system.

[0037] The controller 34 includes at least one processor 44 and a computer readable storage device or medium 46. The processor 44 can be any custom or commercially available processor, a central processing unit (CPU), a graphics processing unit (GPU), a secondary processor in a plurality of processors associated with the controller 34, a semiconductor-based microprocessor (in the form of a microchip or chipset), a macroprocessor, any combination thereof, or any device generally used to execute instructions. The computer readable storage device or medium 46 can include volatile and non-volatile storage in, for example, read-only memory (ROM), random access memory (RAM), and keep-alive memory (KAM). KAM is a permanent or non-volatile memory that can be used to store various operating variables when the processor 44 is powered off. The computer readable storage device or medium 46 can be implemented using any of a variety of known storage devices, such as PROMs (programmable read-only memories), EPROMs (electrically programmable read-only memories), EEPROMs (electrically erasable programmable read-only memories), flash memory, or any other electrical, magnetic, optical, or combination storage device capable of storing data, some of which represent executable instructions used by the controller 34 in controlling the autonomous vehicle 10. In various embodiments, the controller 34 is configured to implement anomaly detection systems and methods, as discussed in detail below.

[0038] The instructions of the controller 34 may include one or more separate programs, each of which includes an ordered list of executable instructions for implementing logical functions. When executed by the processor 44, these instructions receive and process signals from the sensor system 28, perform logic, calculations, methods and / or algorithms for automatically controlling components of the vehicle 10, and generate control signals to the actuator system 30, thereby automatically controlling components of the vehicle 10 based on the logic, calculations, methods and / or algorithms. Although Figure 1 Only one controller 34 is shown, and embodiments of the vehicle 10 may include any number of controllers 34 that communicate via any suitable communication medium or combination of communication media and collaborate to process sensor signals, execute logic, calculations, methods and / or algorithms, and generate control signals to control features of the vehicle 10. As described above, Figure 1 All or part of the abnormality detection system 100 is included in the controller 34.

[0039] Reference now Figure 2 , wherein the operating environment of the anomaly detection system 100 is generally shown at 50, which includes a remote transportation system 52, which is associated with and in communication with one or more vehicles 10a-10n, as shown in reference Figure 1In various embodiments, the operating environment 50 also includes one or more user devices 54 that communicate with the vehicles 10a - 10n and / or the remote transportation system 52 via a communication network 56 .

[0040] The communication network 56 supports the required communications between the devices, systems, and components supported by the operating environment 50 (e.g., via tangible communication links and / or wireless communication links). For example, the communication network 56 may include a wireless carrier system 60, such as a cellular telephone system, which includes a plurality of cell towers (not shown), one or more mobile switching centers (MSCs) (not shown), and any other networking components required to connect the wireless carrier system 60 to a land communication system. Each cell tower includes transmit and receive antennas and a base station, and the base stations from different cell towers are connected to the mobile switching center directly or through intermediate equipment such as a base station controller. The wireless carrier system 60 can implement any suitable communication technology, including, for example, digital technologies such as CDMA (e.g., CDMA2000), LTE (e.g., 4G LTE or 5G LTE), GSM / GPRS, or other current or emerging wireless technologies. Other cell tower / base station / mobile switching center arrangements are also possible and can be used with the wireless carrier system 60. For example, base stations and cell towers may be co-located, or they may be remote from each other, each base station may be responsible for a single cell tower, or a single base station may serve various cell towers, or various base stations may be coupled to a single MSC, to name just a few possible arrangements.

[0041] In addition to the wireless carrier system 60, a second wireless carrier system in the form of a satellite communication system 64 may be included to provide one-way or two-way communications with the vehicles 10a-10n. This may be accomplished using one or more communication satellites (not shown) and uplink transmission stations (not shown). One-way communications may include, for example, satellite radio services, where programming content (news, music, etc.) is received by a transmission station, packaged for upload, and then sent to a satellite, which broadcasts the programming to the user. Two-way communications may include, for example, satellite telephone services, which use satellites to relay telephone communications between the vehicle 10 and the station. A satellite telephone may be used in addition to or in lieu of the wireless carrier system 60.

[0042] A land communication system 62 may also be included, which is a traditional land-based telecommunications network connected to one or more landline telephones and connects the wireless operator system 60 to the remote transportation system 52. For example, the land communication system 62 may include a public switched telephone network (PSTN), such as for providing hard-wired telephone, packet-switched data communications, and Internet infrastructure. One or more portions of the land communication system 62 may be implemented using a standard wired network, fiber optic or other optical network, cable network, power line, other wireless networks such as wireless local area network (WLAN), or a network providing broadband wireless access (BWA), or any combination thereof. In addition, the remote transportation system 52 does not need to be connected via the land communication system 62, but may include a wireless telephone device so that it can communicate directly with a wireless network, such as the wireless operator system 60.

[0043] although Figure 2 Only one user device 54 is shown in the figure, and embodiments of the operating environment 50 can support any number of user devices 54, including multiple user devices 54 owned, operated, or otherwise used by one person. Each user device 54 supported by the operating environment 50 can be implemented using any suitable hardware platform. In this regard, the user device 54 can be implemented in any common form, including but not limited to: a desktop computer; a mobile computer (e.g., a tablet, laptop, or netbook computer); a smart phone; a video game device; a digital media player; a piece of home entertainment equipment; a digital camera or video camera; a wearable computing device (e.g., a smart watch, smart glasses, smart clothing); or the like. Each user device 54 supported by the operating environment 50 is implemented as a computer-implemented or computer-based device having the hardware, software, firmware, and / or processing logic required to perform the various techniques and methods described herein. For example, the user device 54 includes a microprocessor in the form of a programmable device, which includes one or more instructions stored in an internal memory structure and is applied to receive binary input to create a binary output. In some embodiments, the user device 54 includes a global positioning system module capable of receiving global positioning system satellite signals and generating global positioning system coordinates based on these signals. In other embodiments, the user device 54 includes cellular communication functionality, so that the device performs voice and / or data communications over the communication network 56 using one or more cellular communication protocols, as discussed herein. In various embodiments, the user device 54 includes a visual display, such as a touch screen graphical display or other display.

[0044] The remote transportation system 52 includes one or more back-end server systems that may be cloud-based, web-based, or resident at a specific campus or geographic location served by the remote transportation system 52. The remote transportation system 52 may be operated by a live advisor, an automated advisor, or a combination of both. The remote transportation system 52 may communicate with user devices 54 and / or vehicles 10a-10n to schedule rides, dispatch vehicles 10a-10n, deliver information, etc., as will be discussed in more detail below.

[0045] It will be appreciated that the subject matter disclosed herein provides certain enhanced features and functionality to what may be considered a standard or baseline vehicle 10 and / or remote transportation system 52. To this end, the vehicle and remote transportation system may be modified, enhanced, or otherwise supplemented to provide additional features of the anomaly detection system 100 disclosed herein.

[0046] As about Figure 3 As shown in more detail and continue to refer to Figure 1 , the anomaly detection system 100 can be implemented as one or more modules, which are configured to execute one or more methods by, for example, a processor. It is understood that Figure 3 The modules shown may be combined and / or further divided to perform the functions or methods described herein. In addition, inputs to the modules may be received from the sensor system 28, from other control modules (not shown) associated with the vehicle 10, from the communication system 36, and / or by Figure 1 The input may be determined / modeled by other submodules (not shown) within the controller 34. In addition, the input may also be pre-processed, such as sub-sampling, noise reduction, normalization, feature extraction, missing data reduction, etc. In various embodiments, the anomaly detection system 100 includes a training module 102, an anomaly detection module 104, a source identification module 106, a group data storage 108, and a parameter data storage 110.

[0047] In various embodiments, the group data store 108 stores group data 112 associated with sensor pairs. For example, vehicle dynamics can be represented by basic physical equations, where sensor pairs are inherently cross-correlated. Associated groups are created based on these physical equations. For example, given the following relationship:

[0048]

[0049]

[0050]

[0051] and

[0052]

[0053] Five relevant groups can be established, including position, velocity, acceleration, heading and yaw rate. These groups are pre-established and stored as group data 112 in the group data storage 108.

[0054] In various embodiments, parameter data store 110 stores parameter data 114 for use by anomaly detection module 104 and source identification module 106. Parameter data 114 includes threshold data associated with each group.

[0055] In various embodiments, the anomaly detection system 100 is configured to operate in two modes, such as, but not limited to, an offline mode and an online mode. For example, during design and prior to manufacturing, the anomaly detection system 100 operates in a training mode, where parameters are determined and stored in the parameter data store 110 for use by the anomaly detection module 104 and the source identification module 106. In another example, during operation of the vehicle 10, the anomaly detection system 100 operates in an online mode, where signal data is continuously processed based on stored parameters from the parameter data store 110.

[0056] In various embodiments, the training module 102 operates in an offline mode and receives anomaly data 116, signal data 118, and group data 112. The signal data 118 may include non-anomalous data. Based on the received data, the training module 102 trains parameters used in the online mode and stores these parameters as parameter data 114 in the parameter data storage 110. In various embodiments, the parameters include thresholds for each sensor group. The training module 102 determines the thresholds by calculating the minimum mean absolute error (MAE) from the signal data 118, generating a threshold value θ for each corresponding related group. 平均 (θ mean ). For example, the signal data 118 may include multiple trips indicated by vehicle identification and trip identification. For each trip, five correlation groups are applied and an estimated signal is calculated. The estimated signal is subtracted from the received signal to obtain a threshold or trip. If the signal data 118 has N trips, then N thresholds can be calculated. The average of all these N thresholds produces the final θ mean .

[0057] Thereafter, the training module 102 calculates the failure / false alarm rate (FPR) for each relevant group. The false alarm rate is calculated as the ratio between the number of negative events that are incorrectly classified as positive and the total number of actual negative events. For example, when there are no (or very few) anomalies injected on a particular sensor, no anomalies are detected and the calculated FPR is small. The training module stores the calculated FPR and threshold in the parameter data storage 110 as parameter data 114.

[0058] In various embodiments, the anomaly detection module 104 operates in an online mode and receives the signal data 120, the group data 112, and the parameter data 114. The signal data 120 is generated as a result of the operation of the vehicle and can be transmitted in messages within the vehicle 10 or in messages sent from other vehicles. Thus, in various embodiments, anomalies can be detected for sensors within the vehicle or for sensors of other vehicles.

[0059] The anomaly detection module 104 evaluates the signal data 120 based on the groups identified by the group data 112 to determine whether there is an anomaly in one of the signals. For example, in various embodiments, the MAE is calculated for all five relevant groups and compared to the threshold θ indicated by the parameter data 114. When the calculated MAE is greater than the threshold θ, the sensor group is marked as anomaly. When the calculated MAE is less than the threshold θ, the sensor group is marked as non-anomalous. After W samples, the samples are smoothed by a rolling average filter with a window size of W, and then the MAE is recalculated and compared to the threshold again to update the label (anomalous or non-anomalous). In order to reduce measurement errors and improve performance, a batch size B is used for evaluation. For example, when only 0.1*B predictions are anomalies in the last B samples, all predictions are marked as non-anomalous.

[0060] In various embodiments, source identification module 106 receives anomaly detection data 122, parameter data 114, and group data 112. When anomaly detection data 122 indicates that an anomaly is found, source identification module 106 processes the received data to determine the source of the detected anomaly. For example, source identification module 106 calculates a prediction condition vector for each sensor group. The average value of When it is less than the training threshold (FPR) indicated by the parameter data 114, the sensor is not abnormal.

[0061] Example: FPRmean = 0.11, PC p =[0,0,1,0,0,0,0,0,0,0], continue.

[0062] when When greater than the training threshold, then the sensor is anomaly and the source detection data 124 is updated to indicate that an anomaly is detected at the sensor associated with the sensor group.

[0063] In various embodiments, source identification module 106 processes each sensor group through a natural iteration order of the associated groups. For example, source identification module 106 processes each sensor group using three depth-first search (DFS) traversals.

[0064] (1) Position → (2) Velocity → (3) Acceleration → (4) Heading → (5) Yaw rate.

[0065] When the anomaly source is determined, the traversal will end. Therefore, only one anomaly source can be detected. When the entire tree is traversed because the calculated FPR is always below the threshold, no anomaly is detected on any sensor, and the source detection data 124 is updated to indicate that no anomaly is detected.

[0066] Reference now Figure 5 And continue to refer to Figure 1-4 , the flow chart shows that according to the present disclosure, Figure 1 - Method 200, 400 performed by the anomaly detection system 100 of 3. It can be understood from the present disclosure that the order of operations in the method is not limited to the following: Figure 5 In various embodiments, the methods 200 , 400 may be scheduled to run based on one or more predetermined events and / or may run continuously during operation of the vehicle 10 or the remote transportation system 52 .

[0067] In various embodiments, method 200 may be performed to detect anomalies. In one example, method 200 begins at 205. At 207, the values ​​are estimated based on signal data 120 and, for example, the above equations for each correlation group. At 210, the MAE is calculated from all five correlation groups. Then, at 220, the MAE is compared to a threshold value θ indicated by parameter data 114. When the MAE calculated at 220 is greater than the threshold value θ, the sensor group is marked as anomaly at 230, and the anomaly detection data 122 is updated. When the MAE calculated at 220 is less than the threshold value θ, the sensor group is marked as non-anomalous at 240.

[0068] After evaluating W samples at 250, the prediction is updated with the smoothed estimation signal at 260. Thereafter, the prediction count is evaluated at 270. If the prediction count is greater than a defined batch size B at 270, the batch is evaluated at 280. For example, when only a specified percentage of predictions in the last batch of samples are anomalous at 280, all predictions in the batch are marked as non-anomalous at 290. Thereafter, the method continues to monitor the next sample at 210.

[0069] In various embodiments, once an anomaly is detected, method 400 may be performed to detect the source of the anomaly. In one example, method 400 begins at 405. At 410, it is determined that an anomaly is detected. At 420, a sensor group is selected, for example, using the DFS method described above at 420. At 430, a predicted state vector is calculated for the sensor group. The average value is evaluated at 440. When greater than the threshold at 440 , the sensor IS is abnormal and the source detection data 124 is updated to indicate the abnormality at 460 . Thereafter, the method may end at 470 .

[0070] At 440, when When the FPR is less than the threshold value (FPR), the sensor is not abnormal at 450. The method continues to select the next sensor group at 420. If all sensor groups have been processed at 420 and no abnormalities are found, the method can end at 470.

[0071] Although at least one exemplary embodiment has been given in the foregoing detailed description, it should be understood that there are a large number of variations. It should also be understood that one or more exemplary embodiments are only examples and are not intended to limit the scope, applicability or configuration of the present disclosure in any way. On the contrary, the foregoing detailed description will provide a convenient roadmap for those skilled in the art to implement one or more exemplary embodiments. It should be understood that various changes may be made to the functions and arrangements of elements without departing from the scope of the present disclosure set forth in the attached claims and their legal equivalents.

Claims

1. A method for detecting an abnormality of a sensor of a vehicle, comprising: storing a plurality of sensor correlation groups based on vehicle dynamics, wherein each of the plurality of sensor correlation groups includes at least two sensors correlated based on vehicle dynamics; processing a subset of the signals based on at least two sensors in a related group of sensors; processing the subset of signals by calculating a mean absolute error (MAE) based on a correlation group of the plurality of sensors to determine when an anomaly exists; processing a subset of the signals based on the sensor correlation group to determine which sensor in the sensor correlation group is abnormal; and generating notification data based on a sensor of an anomaly in a related group of sensors; Wherein processing the subset of signals to determine which sensor in the sensor correlation group is abnormal is based on a depth first search method and the plurality of sensor correlation groups. 2 . The method of claim 1 , wherein the plurality of sensor-related groups include a position group, a velocity group, an acceleration group, a heading group, and a yaw rate group. 3 . The method of claim 2 , wherein each of the plurality of sensor correlation groups includes one or more sensors that are correlated with the plurality of sensor correlation groups based on vehicle dynamics. 4 . The method of claim 3 , wherein the subset of signals includes a position signal, a velocity signal, an acceleration signal, a heading signal, and a yaw rate signal. 5 . The method of claim 1 , wherein processing the subset of signals to determine which sensor in the sensor correlation group is abnormal is based on a calculated average of predicted condition vectors associated with sensor correlation groups in the plurality of sensor correlation groups. 6 . The method of claim 1 , further comprising training a plurality of thresholds based on abnormal signal data, and wherein processing the subset of signals to determine which sensor in the correlated group of sensors is abnormal is based on the trained plurality of thresholds.

7. The method of claim 1 further comprising training a plurality of thresholds based on abnormal signal data, and wherein processing the subset of signals to determine when a sensor of the sensor-dependent set is abnormal is based on the trained plurality of thresholds.

8. A computer-implemented system comprising: An anomaly detection module comprising one or more processors configured by programming instructions encoded in a non-transitory computer readable medium, the anomaly detection module being configured to: storing a plurality of sensor correlation groups based on vehicle dynamics, wherein each of the plurality of sensor correlation groups includes at least two sensors correlated based on vehicle dynamics; processing a subset of the signals based on at least two sensors in a related group of sensors; processing the subset of signals by calculating a mean absolute error (MAE) based on a correlation group of the plurality of sensors to determine when an anomaly exists; processing a subset of the signals based on the sensor correlation group to determine which sensor in the sensor correlation group is abnormal; and generating notification data based on a sensor of an anomaly in a related group of sensors; Wherein processing the subset of signals to determine which sensor in the sensor correlation group is abnormal is based on a depth first search method and the plurality of sensor correlation groups.

Images