Method and system for identifying an abnormal heat state of a vehicle

Abstract

Method and system are pre3sented for identifying an abnormal heat state of at least one vehicle. The method comprises: receiving a stream of images that comprises a recording of Infrared (IR) radiation of one or more vehicles; identifying at least one single vehicle from said one or more vehicles in said stream of images; determining IR profile of said at least one single vehicle; and analyzing said IR profile of the single vehicles to identify an anomaly indicative of said abnormal heat state in at least one of said at least one single vehicle.

Description

[0001] METHOD AND SYSTEM FOR IDENTIFYING AN ABNORMAL HEAT STATE OF A VEHICLE

[0002] TECHNOLOGICAL FIELD AND BACKROUND

[0003] The present disclosure is in the field of safety of vehicles, in particular in the field of vehicles monitoring for identification of overheating initiation process.

[0004] Overheating is one of serious issues for vehicles. Overheating, if not properly detected and addressed, may cause damage to the engine. Typically, the heat state of a vehicle is controlled by using various temperature sensors installed in the vehicle to monitor the temperature and prevent overheating by triggering alerts or automatic responses.

[0005] GENERAL DESCRIPTION

[0006] The present disclosure provides a system and a method for identifying an abnormal heat state in a vehicle, typically an electric vehicle. The system and the method of the present disclosure make use of IR radiation data that comprises a stream of IR radiation images of vehicles. The IR radiation is indicative of the heat that is emitted from the vehicles and therefore indicative of the temperature of the vehicles or parts thereof. By analyzing the IR profile of the vehicles, signatures of an abnormal heat state can be identified to allow providing an alert and to avoid injuries and damages.

[0007] Therefore, an aspect of the present disclosure provides a method for identifying an abnormal heat state of at least one vehicle. The method comprises: receiving a stream of images that comprises recording(s) of Infrared (IR) radiation originated at one or more vehicles, namely a stream of images being received over a period of time. This includes a continuous IR video that comprises the vehicle. The stream of images can be continuous or non-continuous and can be received from different sensing devices from different locations. For example, the stream of images can be taken from a camera that continuously images a large parking lot with a lot of vehicles in its field of view. In another example, the stream of images can be received from two or more cameras, at least one camera acquiring images in a first checkpoint that the vehicle is passing through and at least one another camera acquiring images in a second checkpoint. It is to be noted that the term "IR radiation" can be expanded, in some embodiments of the method, to thermal imaging. The method further comprises identifying at least one vehicle or single vehicles (it is to be noted that any reference to single vehicles throughout the application should be interpreted as at least one single vehicle) from said one or more vehicles in said stream of images. The term "single vehicles" refers to each isolated entity that is identified as a vehicle, thereby differentiating between different vehicles over time in the stream of images. It is to be noted that the image may include only one vehicle, and the vehicle is recognized and can be correlated with a different identified vehicle in a different image that can be taken at a different checkpoint by different sensing or imaging device(s). The identification is made either based on the image or based on meta data of the image, such as the location where the image was taken, an input of a plate number of the vehicle or any other data that can identify the specific vehicle The method further comprises determining IR profile of said single vehicles and analyzing said IR profile of the single vehicles to identify an anomaly indicative of said abnormal heat state in at least one of said single vehicles. When an abnormal heat state is identified, the method may further comprise generating alert data indicative of the identified abnormal heat state. The alert data may comprise information of the location of the vehicle, the type and characteristics of the abnormal heat state, measures to be taken to deal with the situation, or any combination thereof.

[0008] It is to be noted that any combination of the described embodiments with respect to any aspect of this present disclosure is applicable. In other words, any aspect of the present disclosure can be defined by any combination of the described embodiments.

[0009] In some embodiments of the method, said recording of IR radiation of one or more vehicles is a temporal recording of IR radiation, namely a recording that is taken in two or more time points), and said IR profile is a temporal IR profile, namely the IR profile of the vehicle in two or more time points.

[0010] In some embodiments, the method further comprises recording said recording of IR radiation of the one or more vehicles. The recording may be performed by one or more IR imaging units that are in different locations.

[0011] In some embodiments of the method, said determining comprises determining IR profile of different parts of the vehicle comprising at least a first part and a second part; and wherein said analyzing comprises profiling the IR radiation profile relation between said first part of the vehicle and at least said second part of the vehicle. This can be, for example, a relation between the undercarriage portion of the vehicle or any battery -related part of the vehicle and another part of the vehicle.

[0012] In some embodiments of the method, said determining comprises determining IR profile of different parts of battery pack-related parts of the vehicle.

[0013] In some embodiments of the method, said determining comprises determining IR profile of vehicle parts that are in thermal association with the vehicle's battery pack, such as the undercarriage part of the vehicle, the vehicle's cabin, etc. The term "thermal association" should be understood as parts that can exchange heat between them such that the temperature of one part affects the temperature of the other part.

[0014] In some embodiments of the method, said determining comprises determining the IR profile of the surrounding of said single vehicles. The reference to the surrounding of the vehicle should be understood as the area around the physical boundary of the vehicle that may be affected by thermal radiation from the vehicle.

[0015] In some embodiments of the method, said analyzing comprises analyzing said IR profile of the single vehicles in a selected time window.

[0016] In some embodiments of the method, said selected time window is associated with a time following parking of the vehicle.

[0017] In some embodiments of the method, said selected time window is associated with a time following movement of the vehicle. This can be by towing the vehicle or any self, independent movement of the vehicle.

[0018] In some embodiments of the method, said analyzing comprises comparing the IR profile of a first single vehicle with a second vehicle or a plurality of second vehicles. Namely, the first vehicle is compared with IR profile of many other vehicles that may serve as a reference to determine whether the IR profile of the first vehicle is normal or not. The first vehicle may be the same type or different type than the second vehicle.

[0019] In some embodiments of the method, said analyzing comprises comparing the IR profile of a first single vehicle from said at least one single vehicle with a second IR profile of one or more second vehicles or a model data representing said second IR profile of said one or more second vehicles. The first single vehicle and the one or more second vehicles share at least one characteristic. The characteristic may be any one of the following: the type of the vehicle, the model of the vehicle, the environment that the vehicles are found in, the location that the vehicles are in, the manufacturer, or any combination thereof. Said anomaly is identified if the comparison between the IR profile of the first single vehicle and the second IR profile satisfies a predefined condition. In other words, the method comprises applying a comparison test between the IR profile of the single vehicle and another vehicle, which is either located and identified in the general surrounding of the first vehicle or the information of the second vehicle is predetermined and stored in a database and the data is extracted as the second vehicle shares at least one characteristic with the first vehicle, which can be the same type of vehicle, the same manufacturer, the IR profile of the second vehicle was obtained in similar environmental conditions of the first vehicle, the vehicles are both located in the same space, or any other character. It is assumed that the same vehicle of the same type from the same manufacturer should have the same or similar thermal profile in a similar environment. On this basic assumption, the comparison is performed between the first vehicle thermal, IR profile and the expected thermal profile of the same vehicle type represented by the second vehicle or a model thereof. The comparison may be direct comparison or using ML for this task. The predefined condition may be a difference of temperatures between one or more parts of the first vehicle and the one or more second vehicles, a difference in the temporal IR profile (e.g. the time gradient of the IR profile), a difference in relations between the IR profile of different parts of the vehicles, etc.

[0020] In some embodiments of the method, said first vehicle is the same type of said second vehicle or a plurality of second vehicles.

[0021] In some embodiments of the method, said comparing comprises comparing at least one of a temporal IR profile or an IR profile of similar vehicle's parts. Namely, the analysis comprises identifying whether the IR profile of the first vehicle over time is different than other reference vehicles or whether the IR profile of vehicle parts of the first vehicle is different than similar parts of other reference vehicles.

[0022] In some embodiments of the method, said anomaly is a temporal IR signature. This can be the change rate of the temperature of the vehicle or parts thereof.

[0023] In some embodiments of the method, said abnormal heat state is associated with a battery of the vehicle. Explosions of Lithium-based batteries can cause serious damage to the car and its surroundings and therefore early identification of abnormal condition of the battery can prevent serious hazards.

[0024] In some embodiments of the method, said abnormal heat state is a potential overheating or fire related initiation process in the vehicle or parts thereof, e.g. its battery. This can include actual fire, gas release (including toxic gases), explosions, jet-like flames, etc. The abnormal heat state can be the result of a Thermal Runaway that occurs in the battery.

[0025] In some embodiments of the method, said receiving comprises receiving said stream of images from a single source continuously.

[0026] In some embodiments of the method, said receiving comprises receiving said stream of images from two or more sources continuously, namely two sources that record the stream of images. The two sources can have a common field of view or not.

[0027] In some embodiments of the method, said receiving comprises receiving said stream of images from two or more sources, each source is located so as to record one or more vehicles in a different location and different time point than other sources. For example, one source can record images of the vehicle before it boards a ship for shipping and another source can record the vehicle in its parking aboard the ship. Therefore, the analysis for identifying the anomaly indicative of abnormal state of the vehicle can be based on integration of data collected from multiple sites and / or multiples time stamps of the same single vehicle.

[0028] In some embodiments, the method further comprises outputting an alert indicative of said abnormal heat state, upon identification of said anomaly.

[0029] In some embodiments, the method further comprises detecting a physical damage in one or more parts of a single vehicle from said single vehicles; wherein said analyzing comprises correlating said identified physical damage with the IR profile of the single vehicle to identify said anomaly.

[0030] In some embodiments of the method, said stream of images comprises images of the undercarriage of said one or more vehicles or a ground portion that is thermally associated with the undercarriage of said one or more vehicles.

[0031] In some embodiments, the method further comprises differentiating between electric vehicles and internal combustion engine vehicles (ICE). In some embodiments of the method, said determining and said analyzing is performed only for electric vehicles.

[0032] In some embodiments, the method further comprises environmental data that comprises temperature data of the surrounding of the vehicles. Said analyzing is further based on the environmental data.

[0033] In some embodiments, the method further comprises receiving vehicle data indicative of the vehicle type and its IR standard behavior. Said analyzing is further based on the vehicle data.

[0034] In some embodiments, the method further comprises segmenting said single vehicles into segmented parts; wherein said determining comprises determining IR profile of each of the segmented parts. Said analyzing comprises analyzing each of the segmented parts to identify a heat anomaly in at least one of the following: (i) each of the segmented parts, (ii) relation between at least two of the segmented parts, or (iii) each of the segmented parts and relation between at least two of the segmented parts.

[0035] In other words, the method comprises segmenting the vehicle into segmented parts or sub-parts, each sub-part carries a significant heat data that may indicate that the vehicle undergoes an abnormal heat process. The method comprises analyzing the heat map, which may be also a temporal heat map, of each segmented part. The heat map of each sub-part or the relation between heat maps of two different sub-parts may indicate an abnormal heat state of the vehicle. The segmentation may be performed by image processing, machine learning or Al algorithm. For example, this can be performed based on object detection methods using Single Shot MultiBox Detection (SSD) and Deep Neural Network (DNN). Therefore, the method comprises generating segment IR profile for each segment of the vehicle and analyzing each segment independently and the relations between each segment and one or more of other segments that are thermally associated with it for identifying said anomaly.

[0036] In some embodiments of the method, said segmented parts comprise at least one of: wheels, battery pack zone, battery cell, ground reflection zone. It is to be noted that the ground reflection zones associated with the single vehicle are considered as a vehicle segmented part or sub-part of the vehicle. In some embodiments of the method, said segmented parts comprise two or more battery cells; wherein said analyzing comprises comparing IR profile of a first battery cell with a second battery cell of said two or more battery cells to identify an abnormal difference between the first battery cell and the second battery cell being said anomaly indicative of said abnormal heat state. The different battery cells can be segmented, for example, based on pre-prepared segmentation models or based on battery pack data indicative of the specific structure of the battery pack of the at least one single vehicle, which can be obtained from the manufacturer. By performing the comparison between two battery cells, an identification of a battery cell that is abnormal relative to its neighboring cells within the battery pack can be made. This analysis may assist in preventing Thermal Runaway by alarming battery cells that have an abnormal performance or abnormal IR or thermal profile.

[0037] In some embodiments of the method, said analyzing comprises applying abnormal heat classification for determining whether a heat signature identified ins said IR profile is related to said abnormal heat state or a normal state. For example, heat signature derived from an ICE of a vehicle that is either operating or has been operating in a time frame close to the inspection time is standard, while similar heat identified in an electrical vehicle is abnormal. The abnormal heat classification may be using direct inspection or classification DNNs. The heat emission can be monitored directly or using classification networks in order to improve heat abnormalities inspection. The classification of the state can be single class classification or multiclass classification using classification networks as known in the art. Heat map classification can be either following a segmentation of different parts of the vehicle or without it using direct classification of the heat map.

[0038] In some embodiments of the method, said analyzing comprises detecting structural abnormalities of at least a portion of a battery pack of said at least one single vehicle; wherein the identification of said anomaly indicative of said abnormal heat state is based on the detection of said structural abnormalities. Namely, the anomaly can be identified solely based on the detection of said structural abnormalities or based on a combination of findings of structural abnormalities and heat profile of one or more parts of the vehicle. Therefore, part of the inspection (in particular at checkpoints) may use Al techniques for structural analysis of the vehicle bottom side, which may indicate possibility of Thermal Runaway failure. These structural modifications or abnormalities and related Thermal Runaway possibilities can be monitored by direct inspection algorithms or using DNNs. Optionally, this can be done by comparing the vehicle to history images of the same vehicle, or by comparing the image to a same vehicle of the same type (i.e. standard shape).

[0039] Yet another aspect of the present disclosure provides a system for identifying an abnormal heat state of at least one vehicle. The system comprises a processing circuitry configured for: (i) receiving a stream of images that comprises a recording of Infrared (IR) radiation of one or more vehicles, namely a stream of images over a period of time. This includes a continuous IR video that comprises the vehicle. The stream of images can be continuous or non-continuous and can be received from different sensing devices from different locations. For example, the stream of images can be taken from a camera that continuously images a large parking lot with a lot of vehicles in its field of view. In another example, the stream of images can be received from two cameras, one in a first checkpoint that the vehicle passing through and another in a second checkpoint. It is to be noted that the term "IR radiation" can be expanded, in some embodiments of the method, to thermal imaging; (ii) identifying at least one single vehicle or single vehicles from said one or more vehicles in said stream of images. The term "single vehicles" refers to each isolated entity that it identified as a vehicle, thereby differentiating between different vehicles over time in the stream of images. It is to be noted that the image may include only one vehicle and the vehicle is recognized and can be correlated with a different identified vehicle in a different image that can be taken at a different checkpoint by a different sensing device; (iii) determining IR profile of said single vehicles; and (iv) analyzing said IR profile of the single vehicles to identify an anomaly indicative of said abnormal heat state in at least one of said single vehicles.

[0040] In some embodiments of the system, said recording of IR radiation of one or more vehicles is a temporal recording of IR radiation, namely recording that is taken in two or more time points, and said IR profile is a temporal IR profile, namely the IR profile of the vehicle in two or more time points.

[0041] In some embodiments, the system further comprises one or more memories coupled to the processing circuitry and storing programming instructions for execution by the at least one processing circuitry for performing (i)-(iv). In some embodiments, the system further comprises one or more IR imaging units for recording said recording of IR radiation of the one or more vehicles.

[0042] In some embodiments of the system, said determining comprises determining IR profile of different parts of the vehicle comprising at least a first part and a second part; and wherein said analyzing comprises profiling the IR radiation profile relation between said first part of the vehicle and at least said second part of the vehicle. This can be, for example, a relation between the undercarriage portion of the vehicle or any battery -related part of the vehicle and another part of the vehicle.

[0043] Therefore, in some embodiments of the system, said determining comprises determining IR profile of vehicle parts that are in thermal association with the vehicle's battery pack, such as the undercarriage part of the vehicle, the vehicle's cabin, etc. The term "thermal association" should be understood as parts that can exchange heat between them such that the temperature of one part affects the temperature of the other part.

[0044] In some embodiments of the system, said determining comprises determining IR profile of different parts of battery pack-related parts of the vehicle.

[0045] In some embodiments of the system, said determining comprises determining the IR profile of the surroundings of said single vehicles. The reference to the surroundings of the vehicle should be understood as the area around the physical boundary of the vehicle that may be affected by thermal radiation from the vehicle.

[0046] In some embodiments of the system, said analyzing comprises analyzing said IR profile of the single vehicles in a selected time window.

[0047] In some embodiments of the system, said selected time window is associated with a time following parking of the vehicle.

[0048] In some embodiments of the system, said selected time window is associated with a time following movement of the vehicle. This can be by towing the vehicle or any self, independent movement of the vehicle.

[0049] In some embodiments of the system, said analyzing comprises comparing the IR profile of a first single vehicle with a second vehicle or a plurality of second vehicles. Namely, the first vehicle is compared with IR profile of many other vehicles that may serve as a reference to determine whether the IR profile of the first vehicle is normal or not. The first vehicle may be the same type or different type than the second vehicle. In some embodiments of the system, said analyzing comprises comparing the IR profile of a first single vehicle from said at least one single vehicle with a second IR profile of one or more second vehicles or a model representing said second IR profile of said one or more second vehicles, wherein the first single vehicle and the one or more second vehicles share at least one characteristic. The characteristic may be any one of the following: the type of the vehicle, the environment that the vehicles are found in, the location that the vehicles are in, the manufacturer, or any combination thereof. Said anomaly is identified if the comparison between the IR profile of the first single vehicle and the second IR profile satisfies a predefined condition. A comparison between two vehicles sharing similar characteristics may indicate that there is an anomaly in one of them. Therefore, making this comparison facilitates identifying anomalies.

[0050] In some embodiments of the system, said first vehicle is the same type as said second vehicle or a plurality of second vehicles.

[0051] In some embodiments of the system, said comparing comprises comparing at least one of a temporal IR profile or an IR profile of similar vehicle's parts. Namely, the analysis comprises identifying whether the IR profile of the first vehicle over time is different than other reference vehicles or whether the IR profile of vehicle parts of the first vehicle is different than similar parts of other reference vehicles.

[0052] In some embodiments of the system, said anomaly is a temporal IR signature. This can be the change rate of the temperature of the vehicle or parts thereof.

[0053] In some embodiments of the system, said abnormal heat state is associated with a battery of the vehicle.

[0054] In some embodiments of the system, said abnormal heat state is a potential overheating or fire related initiation process in the vehicle or parts thereof, e.g. its battery. This can include actual fire, gas release (including toxic gases), explosions, jet-like flames, etc. The abnormal heat state can be the result of a Thermal Runaway that occurs in the battery.

[0055] In some embodiments of the system, said receiving comprises receiving said stream of images from a single source continuously. In some embodiments of the system, said receiving comprises receiving said stream of images from two or more sources continuously, namely two sources that record the stream of images. The two sources can have a common field of view or not.

[0056] In some embodiments of the system, said receiving comprises receiving said stream of images from two or more sources, each source is located so as to record one or more vehicles in a different location and different time point than other sources. For example, one source can record images of the vehicle before it boards a ship for shipping and another source can record the vehicle in its parking aboard the ship.

[0057] In some embodiments of the system, the processing circuitry is configured to generate, upon identification of said anomaly, alert data indicative of said abnormal heat state.

[0058] In some embodiments, the system further comprises an output module that is configured to output said alert data.

[0059] In some embodiments of the system, the processing circuitry is configured for detecting physical damage in one or more parts of a single vehicle from said single vehicles, wherein the detection is based on the IR data. Said analyzing comprises correlating said identified physical damage with the IR profile of the single vehicle to identify said anomaly.

[0060] In some embodiments of the system, said stream of images comprises images of the undercarriage of said one or more vehicles or a ground portion that is thermally associated with the undercarriage of said one or more vehicles.

[0061] In some embodiments of the system, the processing circuitry is configured for differentiating between electric vehicles and internal combustion engine vehicles.

[0062] In some embodiments of the system, said determining and said analyzing is performed only for electric vehicles.

[0063] In some embodiments of the system, the processing circuitry is configured to receive environmental data that comprises temperature data of the surrounding of the vehicles. Said analyzing is further based on the environmental data. In some embodiments of the system, the processing circuitry is configured to receive vehicle data indicative of the vehicle type and its IR standard behavior. Said analyzing is further based on the vehicle data.

[0064] In some embodiments of the system, the processing circuitry is configured for segmenting said single vehicles into segmented parts. Said determining comprises determining IR profile of each of the segmented parts. Said analyzing comprises analyzing each of the segmented parts to identify a heat anomaly in at least one of the following: (i) each of the segmented parts, (ii) relation between at least two of the segmented parts, or (iii) each of the segmented parts and relation between at least two of the segmented parts. The segmentation may be performed by machine learning or Al algorithm. For example, this can be performed based on object detection methods using Single Shot MultiBox Detection (SSD) and Deep Neural Network (DNN).

[0065] In some embodiments of the system, said segmented parts comprise at least one of: wheels, battery pack zone, battery cell, ground reflection zone.

[0066] In some embodiments of the system, said segmented parts comprise two or more battery cells; wherein said analyzing comprises comparing IR profile of a first battery cell with a second battery cell of said two or more battery cells to identify an abnormal difference between the first battery cell and the second battery cell being said anomaly indicative of said abnormal heat state.

[0067] In some embodiments of the system, said analyzing comprises applying abnormal heat classification for determining whether a heat signature identified ins said IR profile is related to said abnormal heat state or a normal state.

[0068] In some embodiments of the system, said analyzing comprises detecting structural abnormalities of at least a portion of a battery pack of said at least one single vehicle; wherein the identification of said anomaly indicative of said abnormal heat state is based on the detection of said structural abnormalities. The anomaly can be identified solely based on the detection of said structural abnormalities or based on a combination of findings of structural abnormalities and heat profile of one or more parts of the vehicle. Therefore, part of the inspection (in particular at checkpoints) may use Al techniques for structural analysis of the vehicle bottom side, which may indicate possibility of Thermal Runaway failure. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] In order to better understand the subject matter that is disclosed herein and to exemplify how it may be carried out in practice, embodiments will now be described, by way of non-limiting examples only, with reference to the accompanying drawings, in which:

[0070] Fig. 1 is a block diagram of a non-limiting example of an embodiment of the system for identifying abnormal heat state in a vehicle according to an aspect of the present disclosure.

[0071] Fig. 2 is a flow diagram of a method for identifying an abnormal heat state of at least one vehicle.

[0072] Fig. 3 is a schematic illustration exemplifying two modes of monitoring vehicles, intermittent monitoring and surveillance monitoring.

[0073] Fig. 4 is a schematical chart exemplifying the temperature change along the timeline of failure sequence of a battery cell of an electrical vehicle.

[0074] Fig. 5 is a schematic illustration exemplifying the causes for thermal runaway.

[0075] DETAILED DESCRIPTION

[0076] The following figures are provided to exemplify embodiments and realization of the invention of the present disclosure.

[0077] Reference is first being made to Fig. 1, which is a block diagram of a non-limiting example of an embodiment of the system for identifying abnormal heat state in a vehicle, typically an electric vehicle, according to an aspect of the present disclosure. The system 100 comprises an input module 102 that is configured to receive IR radiation data IRD that comprises a stream of IR images that comprises recording of IR radiation. The source of the IR radiation data IRD can be one or a plurality of cameras that are positioned so as to provide a recording of either a single vehicle at a time, e.g. a checkpoint camera, or a plurality of vehicles at a time, e.g. a camera that is positioned in parking lot so as to capture simultaneously many vehicles in a surveillance mode. These two options are exemplified in Fig. 3. Checkpoint imaging is regarded as intermittent monitoring (the upper part of Fig. 3) and surveillance imaging is regarded as continuous monitoring (the bottom part of Fig. 3). The input module 102 may optionally further receive vehicle data VD that comprises information of the specific vehicle type that is identified. The vehicle data VD may include standard IR profile behavior for comparison. The standard IR profile behavior may be based on the history of monitoring of a plurality of similar vehicle types. The input module 102 may optionally further receive environmental data ED indicative of the environmental conditions of the surrounding of the imaged vehicles, such as the surrounding temperature, humidity, etc. The system 100 further comprises a processing circuitry 104 that is configured for receiving said IR radiation data IRD and identifying single vehicles. Further to the identification of single vehicles, the processing circuitry 104 may be configured for differentiating between an electric vehicle and internal combustion engine vehicles. Since the hazard of an abnormal heat state of an electric vehicle is much greater, due to the risk of the explosion of the Lithium battery, the processing circuitry may focus primarily or solely in the identified single electric vehicles. Therefore, in order to be able to monitor an abnormal heating process in a vehicle, the processing circuitry isolates each vehicle and monitors the IR profile of the vehicle, typically the IR temporal profile of the vehicle. The processing circuitry 104 is further configured to analyze the IR profile of the vehicle to identify signatures indicative of an abnormal heat state of the vehicle and generate an alert data AD indicative of the abnormal heat state of the specific vehicle in which the abnormal heat state is identified. The processing circuitry 104 may optionally receive the vehicle data VD and the environmental data ED and perform the analysis of the IR profile also based on these data. This means that the analysis includes comparison of the specific IR profile of an imaged vehicle with the vehicle data VD and taking into account the surrounding environmental conditions. The alert data AD may comprise information of the type of abnormal heat state that is identified and / or the specific position of the vehicle. Furthermore, the alert data AD may optionally include information about the suggested measures that should be taken in order to manage the situation. This may include evacuation around the vehicle and types of fire extinguishing tools to be used in the situation. An output module 106 of the system 100 is configured to receive the alert data AD and transmit it to any necessary recipient, e.g. security team that in charge of the parking place of the vehicle. It is to be noted that even the input module 102 and the output module 106 can be independent components as exemplified in Fig. 1, they can also be part of the processing circuitry 102.

[0078] Reference is now being made to Fig. 2, which is a flow diagram of a method for identifying an abnormal heat state of at least one vehicle, typically an electric vehicle. The method comprises receiving IR radiation data that comprises IR recording images 250. The source of the IR radiation data can be one or a plurality of cameras that are positioned so as to provide a recording of either a single vehicle at a time, e.g. a checkpoint camera, or a plurality of vehicles at a time, e.g. a camera that is positioned in parking lot so as to capture simultaneously many vehicles in a surveillance mode. These two options are exemplified in Fig. 3.

[0079] Checkpoint imaging is regarded as intermittent monitoring (the upper part of Fig. 3) and surveillance imaging is regarded as continuous monitoring (the bottom part of Fig. 3). The method further comprises identifying single vehicles in the stream of images 252.

[0080] The identification can further comprise tagging each vehicle as being either an electric vehicle or an internal combustion engine vehicle. This is important as the IR behavior of each vehicle type is very different and also the hazard. Explosion of an electric vehicle while in parking is more common and much more dangerous and destructive. Since the temperatures of electric vehicles are typically much lower than the internal combustion engine vehicles, during driving or right after parking of the vehicle, when the vehicle enters the line of sight of the IR source that provides the IR recording, it can be easily tagged.

[0081] The method further comprises determining IR profile of the identified single vehicles 254. The IR profile can be either spatial (namely the profile of different parts of the vehicle and the IR profile of the surrounding of the vehicle), temporal or a combination of spatial and temporal profiles.

[0082] The method further comprises analyzing said IR profile to identify abnormal heat state of the vehicle 256. The analysis can be based on either the temporal or the spatial IR behavior of the vehicle. This can be based only on the IR profile of the vehicle itself and / or based on comparison with other vehicles. If an abnormal heat state is identified, the method further comprises generating output data indicative of the identified abnormal heat state of the vehicle 258. The output data can be transmitted to any desired entity that is in charge of the vehicle or the space where the vehicle parks in.

[0083] Fig. 4 is a schematic chart exemplifying the temperature change during a failure sequence of a battery or a battery cell of an electrical vehicle. The curved arrow showing the rise of the temperature (the vertical axis) vs the timeline with the indication of each step of the failure and the range of temperatures associated with the failure step. The sequence is reversible until the step of solid electrolyte interphase (SEI) decomposition.

[0084] The purpose of the present disclosure is to provide a solution that can identify the sequence on its early, reversible steps in order to prevent from reaching to a thermal runaway. However, even if the sequence is detected in the non-reversible steps, the present disclosure provides a solution for identifying it and alert about it so safety measure can be taken.

[0085] There are several reasons that may cause thermal runaway, such as mechanical abuse, electrical abuse, thermal abuse, and battery aging. These are described for example by Billy Wu in his publication of May 1, 2023 in BatteryDesign.net / rom chemistry to pack website, which is incorporated herein by reference. Fig. 5 schematically exemplifies, in a self-explanatory manner, the reasons leading for thermal runaway. As shown in Fig. 5, in addition to the reasons described in the above-indicated article, manufacturing defects and others may also affect thermal runaway.

Claims

CLAIMS:

1. A method for identifying an abnormal heat state of at least one vehicle, comprising: receiving a stream of images that comprises a recording of Infrared (IR) radiation of one or more vehicles; identifying at least one single vehicle from said one or more vehicles in said stream of images; determining IR profile of said at least one single vehicle; analyzing said IR profile of the single vehicles to identify an anomaly indicative of said abnormal heat state in at least one of said at least one single vehicle.

2. The method of claim 1, wherein said recording of IR radiation of one or more vehicles is a temporal recording of IR radiation, and said IR profile is a temporal IR profile.

3. The method of claim 1 or 2, comprising recording said recording of IR radiation of the one or more vehicles.

4. The method of any one of claims 1-3, wherein said determining comprises determining IR profile of different parts of the vehicle comprising at least a first part and a second part; and wherein said analyzing comprises profiling an IR radiation profile relation between said first part of the vehicle and at least said second part of the vehicle.

5. The method of any one of claims 1-4, wherein said determining comprises determining IR profile of different parts of battery pack-related parts of the vehicle.

6. The method of any one of claims 1-5, wherein said determining comprises determining the IR profile of the surrounding of said at least one single vehicle.

7. The method of any one of claims 1-6, wherein said analyzing comprises analyzing said IR profile of the at least one single vehicle in a selected time window.

8. The method of claim 7, wherein said selected time window is associated with a time following parking of the at least one single vehicle.

9. The method of claim 7, wherein said selected time window is associated with a time following movement of the vehicle.

10. The method of any one of claims 1-9, wherein said analyzing comprises comparing the IR profile of a first single vehicle from said at least one single vehicle with a second IR profile of one or more second vehicles or a model representing said second IR profile of said one or more second vehicles, wherein the first single vehicle and the one or more second vehicles share at least one characteristic; wherein said anomaly is identified if the comparison between the IR profile of the first single vehicle and the second IR profile satisfies a predefined condition.

11. The method of claim 10, wherein said at least one characteristic comprises at least one of the following: a type of the vehicle, a model of the vehicle, environment that the vehicles are found in, a location that the vehicles are in, and a type or manufacturer.

12. The method of claim 10 or 11, wherein said comparing comprises comparing at least one of a temporal IR profile or an IR profile of similar vehicle's parts.

13. The method of any one of claims 1-12, wherein said anomaly is a temporal IR signature.

14. The method of any one of claims 1-13, wherein said abnormal heat state is associated with a battery of the vehicle.

15. The method of any one of claims 1-14, wherein said abnormal heat state is a potential fire related initiation process in the vehicle or parts thereof.

16. The method of any one of claims 1-15, wherein said receiving comprises receiving said stream of images from a single source continuously.

17. The method of any one of claims 1-15, wherein said receiving comprises receiving said stream of images from two or more sources continuously.

18. The method of any one of claims 1-15, wherein said receiving comprises receiving said stream of images from two or more sources, each source is located so as to record one or more vehicles in a different location and different time point than other sources.

19. The method of any one of claims 1-18, comprising outputting an alert indicative of said abnormal heat state, upon identification of said anomaly.

20. The method of any one of claims 1-19 comprising detecting a physical damage in one or more parts of a single vehicle from said single vehicles; wherein said analyzingcomprises correlating said identified physical damage with the IR profile of the single vehicle to identify said anomaly.

21. The method of any one of claims 1-20, wherein said stream of images comprises images of the undercarriage of said one or more vehicles or a ground portion that is thermally associated with the undercarriage of said one or more vehicles.

22. The method of any one of claims 1-21, comprising differentiating between electric vehicles and internal combustion engine (ICE) vehicles; wherein said determining and said analyzing is performed only for electric vehicles.

23. The method of any one of claims 1-22, comprising segmenting said single vehicles into segmented parts; wherein said determining comprises determining IR profile of each of the segmented parts; wherein said analyzing comprises analyzing each of the segmented parts to identify a heat anomaly in at least one of the following: (i) each of the segmented parts, (ii) relation between at least two of the segmented parts, or (iii) each of the segmented parts and relation between at least two of the segmented parts.

24. The method of claim 23, wherein said segmented parts comprise at least one of: wheels, battery pack zone, battery cell, ground reflection zone.

25. The method of claim 23 or 24, wherein said segmented parts comprise two or more battery cells; wherein said analyzing comprises comparing IR profile of a first battery cell with a second battery cell of said two or more battery cells to identify an abnormal difference between the first battery cell and the second battery cell being said anomaly indicative of said abnormal heat state.

26. The method of any one of claims 1-25, wherein said analyzing comprises applying abnormal heat classification for determining whether a heat signature identified ins said IR profile is related to said abnormal heat state or a normal state.

27. The method of any one of claims 1-26, wherein said analyzing comprises detecting structural abnormalities of at least a portion of a battery pack of said at least one single vehicle; wherein the identification of said anomaly indicative of said abnormal heat state is based on the detection of said structural abnormalities.

28. A system for identifying an abnormal heat state of at least one vehicle, comprising: a processing circuitry configured for: receiving a stream of images that comprises a recording of Infrared (IR) radiation of one or more vehicles; identifying at least one single vehicle from said one or more vehicles in said stream of images; determining IR profile of said at least one single vehicle; and analyzing said IR profile of the at least one single vehicle to identify an anomaly indicative of said abnormal heat state in at least one of said at least one single vehicle.

29. The system of claim 28, wherein said recording of IR radiation of one or more vehicles is a temporal recording of IR radiation, and wherein said IR profile is a temporal IR profile.

30. The system of claim 28 or 29, comprising one or more IR imaging units for recording said recording of IR radiation of the one or more vehicles.

31. The system of any one of claims 28-30, wherein said determining comprises determining IR profile of different parts of the at least one single vehicle comprising at least a first part and a second part; and wherein said analyzing comprises profiling the IR radiation profile relation between said first part of the at least one single vehicle and at least said second part of the at least one single vehicle.

32. The system of any one of claims 28-31, wherein said determining comprises determining IR profile of different parts of battery pack-related parts of the at least one single vehicle.

33. The system of any one of claims 28-32, wherein said determining comprises determining the IR profile of the surrounding of said at least one single vehicle.

34. The system of any one of claims 28-33, wherein said analyzing comprises analyzing said IR profile of the at least one single vehicle in a selected time window.

35. The system of claim 34, wherein said selected time window is associated with a time following parking of the at least one single vehicle.

36. The system of claim 35, wherein said selected time window is associated with a time following movement of the at least one single vehicle.

37. The system of any one of claims 28-36, wherein said analyzing comprises comparing the IR profile of a first single vehicle from said at least one single vehicle with a second IR profile of one or more second vehicles or a model representing said second IR profile of said one or more second vehicles, wherein the first single vehicle and the one or more second vehicles share at least one characteristic; wherein said anomaly is identified if the comparison between the IR profile of the first single vehicle and the second IR profile satisfies a predefined condition.

38. The system of claim 37, wherein said at least one characteristic comprises at least one of the following: a type of the vehicle, a model of the vehicle, environment that the vehicles are found in, a location that the vehicles are in, and a type or manufacturer.

39. The system of claim 37 or 38, wherein said comparing comprises comparing at least one of a temporal IR profile or an IR profile of similar vehicle's parts.

40. The system of any one of claims 28-39, wherein said anomaly is a temporal IR signature.

41. The system of any one of claims 28-40, wherein said abnormal heat state is associated with a battery of the at least one single vehicle.

42. The system of any one of claims 28-41, wherein said abnormal heat state is a potential fire related initiation process in the at least one single vehicle or parts thereof.

43. The system of any one of claims 28-42, wherein the processing circuitry is configured to generate, upon identification of said anomaly, alert data indicative of said abnormal heat state.

44. The system of any one of claims 28-43, wherein the processing circuitry is configured for detecting a physical damage in one or more parts of a single vehicle from said single vehicles; wherein said analyzing comprises correlating said identified physical damage with the IR profile of the single vehicle to identify said anomaly.

45. The system of any one of claims 28-44, wherein said stream of images comprises images of the undercarriage of said one or more vehicles or a ground portion that is thermally associated with the undercarriage of said one or more vehicles.

46. The system of any one of claims 28-45, wherein the processing circuitry is configured for differentiating between electric vehicles and internal combustion engine vehicles; wherein said determining and said analyzing is performed only for electric vehicles.

47. The system of any one of claims 28-46, wherein the processing circuitry is configured for segmenting said single vehicles into segmented parts; wherein said determining comprises determining IR profile of each of the segmented parts; wherein said analyzing comprises analyzing each of the segmented parts to identify a heat anomaly in at least one of the following: (i) each of the segmented parts, (ii) relation between at least two of the segmented parts, or (iii) each of the segmented parts and relation between at least two of the segmented parts.

48. The system of claim 47, wherein said segmented parts comprise at least one of: wheels, battery pack zone, battery cell, ground reflection zone.

49. The system of claim 47 or 48, wherein said segmented parts comprise two or more battery cells; wherein said analyzing comprises comparing IR profile of a first battery cell with a second battery cell of said two or more battery cells to identify an abnormal difference between the first battery cell and the second battery cell being said anomaly indicative of said abnormal heat state.

50. The system of any one of claims 28-49, wherein said analyzing comprises applying abnormal heat classification for determining whether a heat signature identified ins said IR profile is related to said abnormal heat state or a normal state.

51. The system of any one of claims 28-50, wherein said analyzing comprises detecting structural abnormalities of at least a portion of a battery pack of said at least one single vehicle; wherein the identification of said anomaly indicative of said abnormal heat state is based on the detection of said structural abnormalities.

Images