FIELD OF THE INVENTION

The present invention relates to a method of identifying transport vehicle type of a vehicle carrying an asset having an associated logger device while the asset is transported from an origin location to a destination location.

BACKGROUND OF THE INVENTION

With the expansion and growth of global sourcing in a supply chain, more prevalent interest has been placed on the automatic electronic time and monitoring of environment related parameters to increase food and drug safety and improve food defense systems throughout all areas of production, processing, storage and transportation, and operations. Food and drug require proper handling of environment related parameters such as temperature during transport to assure shelf quality, longevity, safety and effectiveness.

Logger devices are electronic monitoring devices commonly used for these purposes, namely, to be associated with assets such as food, beverages or drugs to automatically monitor and record various environmental related parameters of the assets throughout a supply chain, such as temperature, humidity, acceleration, and air pressure, over time. A recent example of the importance of such logger devices is the temperature monitoring of the COVID-19 vaccines, which is a key critical monitoring parameter.

Today, the logger devices are typically only provided with information relating to origin location (e.g. manufacturing facility) and destination location (e.g. distribution center) of the asset, but other types of details during the supply chain such as the type of transport vehicles are currently not available.

It would be beneficial to be able to automatically identify in what type of a transport means a logger device is currently in, such as if it is onboard of vehicle such as an aircraft, truck, ship, train etc..

Identifying if the logger device is onboard an aircraft is an important safety measure to ensure that the logger device is automatically switched to a low power mode/ airflight mode during flight, instead of relying on manual operation which is commonly done today. On the other hand, identifying the transport vehicle type of a vehicle currently transporting the asset ensures that the logger device is not accidentally switched off during transport where it is needed, where this would result in a data gap during the supply chain.

Further, such a transport identification may link excursions such as temperature excursions to different transport vehicle types so that it may be possible to optimize the supply chain monitoring by identifying where or in what type of transport vehicle such excursions happen more frequently.

US2021385620 discloses an asset logger device and a method of determining whether an logger device is transported by a predetermined type of transport means by means of recording 3 axes acceleration data using an accelerometer while transporting the asset from its location of origin to a destination location. The step of determining the transport type comprises counting by a controller in the tracking device the change of sign of the sensed acceleration along 3 axes during a predetermined amount of time, where from the counted changes it is determined whether the logger device is transported by the predetermined type of transportation. The fact that the above calculations are performed for said 3 axis makes the calculation very complex and requires high processing power by the controller at the cost of the power source in the logger device.

Another problem with the method in US2021385620 is that is must be ensured that the logger devices are placed in the exact same way within the asset (e.g. in a package containing the asset), e.g. horizontal/vertical/flat, otherwise the acceleration data will in 3D space will be different due to different orientation of the accelerometer. This is illustrated in figure 2, showing acceleration data in 3D space, for the same logger device having different orientation in the same aircraft.

It would be beneficial to provide a simplified method where less processing power is needed to identify the transport vehicle type of a vehicle currently transporting an asset having an associated logger device during the supply chain monitoring, and that is independent on the orientation of the placement of the logger devices. Also, with respect to sustainability, by identifying the transport vehicle type it is possible to estimate the carbon footprint during the transportation, and knowing the carbon footprint during the transportation is important when calculating sustainability score which is highly relevant today. This score data can be important to identify the most the optimal route with minimal carbon footprint. SUMMARY OF THE INVENTION

It is an object of the invention to provide a simplified and less power consuming solution in identifying transport vehicle type that is transporting an asset having associated a logger device by means of utilizing acceleration vectors measured by an accelerometer in the logger device while the transport vehicle type is transporting the asset from an origin location to a destination location.

In general, the invention preferably seeks to mitigate, alleviate, or eliminate one or more of the above-mentioned disadvantages of the prior art singly or in any combination. In particular, it may be seen as an object of embodiments of the present invention to provide a method, a logger device and a system that solves the above-mentioned problems, or other problems.

To better address one or more of these concerns, in a first aspect of the invention a method is provided for identifying transport vehicle type of a vehicle which is currently transporting an asset having an associated logger device while the asset is transported from an origin location to a destination location, the method comprising:

• receiving a measured acceleration data from the logger device, where the acceleration data comprises a collection of acceleration vectors,

• transforming the received collection of acceleration vectors into direction-independent acceleration data, and

• utilizing the direction-independent acceleration data as input data in identifying the transport vehicle type of the vehicle which is currently transporting the asset.

Measurements show namely that different orientation of logger devices during use is, e.g. if they are placed horizontally or vertically on or into package containing the asset, reflected in different mapping results of acceleration data. Thus, this problem is overcome by transforming the received collection of acceleration vectors into direction-independent acceleration data. This simplifies the subsequent calculation and therefore the processing power that would otherwise be needed to make adjust the different mapping results.

Accordingly, the method of the present invention eliminates this direction dependency and provides an energy efficient method of automatically identifying transport vehicle types where excursions occur more frequently than for other transport vehicle types during transport in the supply chain. Based thereon this additional information is valuable for planning the transport vehicle types within the supply chain with the aim of minimizing excursion rate. Also, by automatically identifying if a logger device is onboard an aircraft onboard an aircraft instead of relying on manual labour in switching the logger device to airflight mode increases flight safety and eliminates human mistakes such as by forgetting to switch the logger device to airflight mode.

Moreover, it also prevents the logger device from accidentally switching to lower power mode/ airflight mode or switch off when not needed (because it is e.g. in a truck), which would otherwise result in a data gap during the supply chain monitoring.

Further, having information about the transport vehicle type of the vehicle currently transporting the asset, which may be, train, ship, truck or aircraft etc., it is possible to estimate the carbon footprint of the transport vehicle and thus the transport route may be further optimized with respect to minimizing the carbon footprint. This valuable information may be used in assisting customers in changing their transportation routes with the aim of minimizing their carbon footprint.

In one embodiment, the step of transforming the received collection of acceleration vectors into direction-independent acceleration data comprises transforming a chunk sample of acceleration vectors collected over pre-defined time-period, wherein the step of transforming the received collection of acceleration vectors into direction-independent acceleration data results in a single data value, wherein the step of transforming the chunk sample of acceleration vectors collected over the pre-defined time-period is repeated multiple times over a time-period t resulting in a multiple of said data values, where the time evolution of the multiple of data values over the time-period t is compared with pre-defined reference data when identifying the vehicle type currently transporting the asset. The pre-defined reference data may be collected by placing such logger devices in different vehicle types for days, weeks, months until solid “fingerprint” data is obtained that uniquely identifies the vehicle types. For a given vehicle type this may also be done for different drive mode, e.g. stationary vehicle with engine running, the same vehicle driving in a city where the traffic level is above a given reference number of population, the same vehicle driving long distance in a highway etc.. So, for the same vehicle type, there may be several “fingerprint” data for said different drive mode. Thus, the above- mentioned step of comparing the time evolution of the multiple of data values with said predefined reference data when identifying the vehicle type currently transporting the asset is performed by checking for a matching with the reference data. The pre-defined time-period may as an example be, but is not limited to, a fraction of a second or several seconds, where the number of acceleration data collection during this period is processed and results in said single value. As an example, the accumulation of acceleration data for each chunk is Ip second and where this is repeated for example 4000 times, which results in 4000 data points.

In one embodiment, the step of transforming the received collection of acceleration vectors into direction-independent acceleration data comprises calculating amplitude variance for the acceleration vectors, where said single data value is a single amplitude variance value.

In another embodiment, the step of transforming the received collection of acceleration vectors into direction-independent acceleration data comprises computing the norm of the acceleration vectors and subsequently calculating the entropy of the norms of the acceleration vectors, where said single data value is a single entropy value.

Accordingly, a simple solution is provided to eliminate the direction dependency of the input data by converting it into a direction independent data value that yet gives a very good indication about the relevance of the acceleration data.

There are many variants of entropy e.g. approximate entropy (ApEN), sample entropy (SampEN), permutation entropy (PE) etc., and the standard Shannon entropy which may be preferred because of the simplicity to use it. Now, as an example, to calculate the Shannon entropy of the norms of the acceleration vectors for within each chunk sample is divided by the sum of the norm values for the acceleration vectors in the chunk sample which transforms it to a probability distribution p (since the Shannon entropy is defined for a probability distribution, by normalizing the chunk sample it becomes transformed into a probability distribution). Then the Shannon entropy is computed according to the definition H = -sum p(x) * log2(p(x)) where the sum is taken over the norm values of the chunk samples.

In yet another embodiment, the Total Variation (TV) of each chunk of N measurements is calculated for the acceleration along an x-axis, as follows: where this is in similar way calculated for y and z-axis. To have a size independent of rotation the values for each axis can be taken together into one size e.g. by taking the Euclidean-norm of the total variation vector tv = (tv _x , tv _y , tv _z In one embodiment, the step utilizing the direction-independent acceleration data as input in identifying the transport vehicle type of the asset comprises applying at least one filter to the direction-independent acceleration data where the resulting data after applying the at least one filter to the direction-independent acceleration data is processed and utilized in identifying the transport vehicle type. The processing power may thus be reduced by selecting the most relevant filtering window relevant for the different transport vehicle types and data “noise” is eliminated.

In one embodiment, the transport vehicle type of the vehicle is selected from:

• a truck or a car,

• a ship,

• a train,

• an aircraft, wherein each of the transport vehicle types has a direction-independent acceleration data that uniquely characterizes the transport vehicle type.

In one embodiment, the step of transforming the received collection of acceleration vectors into direction-independent acceleration data and utilizing the direction-independent acceleration data as input in identifying the transport vehicle type of the vehicle currently transporting the asset is performed by a processor comprised in the logger device.

In another embodiment, the collection of acceleration vectors is transmitted to an external computer, and where the step of transforming the received collection of acceleration vectors into direction-independent acceleration data and utilizing the direction-independent acceleration data as input in identifying the transport vehicle type of the vehicle currently transporting the asset is performed by the external computer.

Accordingly, depending on e.g. communication quality such as 4G connection quality, there may be no other option than to use the processing power of the processor in the logger device, although that requires rather high battery power in particular when communicating the result to the external computer, whereas in case good communication quality is available the “raw acceleration data” may be communicated to the external computer the performs the processing steps.

In one embodiment, the logger device is configured to measure at least one environment related parameter and communicate the at least one measured environment related parameter together with the position data of the logger device to an external computer by means of temporarily adjusting the power mode of the logger device to a higher energy mode while communication with the external computer. The measured environmental related parameters are commonly temperature of the asset or around the asset, humidity, light intensity, the pressure, the angular position of the logger device, and the like, where this measured data is preferably stored in a memory unit comprised in the logger device. As an example, the temperature may be measured every 10 minutes and stored preferably together with a time stamp.

The communication module comprises in one embodiment a modem where the communication is via a communication network such as 3G, 4G, 5G networks, where during communication the modem is set to a higher energy mode where data transmission is possible. This is however an energy consuming process meaning that this higher energy state is only activated temporarily during the data transmission, which may e.g. be once every hour, or once every 2 hours, where the transmission commonly takes less than a minute or several minutes, depending on the signal quality of the communication network. Subsequently, the modem is set to a low power mode where such data transmission is not possible to ensure minimal power consumption of the logger device.

In one embodiment, the step of transforming the received collection of acceleration vectors into direction-independent acceleration data and utilizing the direction-independent acceleration data as input in identifying the transport vehicle type of the vehicle currently transporting the asset is performed by a processor comprised in the logger device, wherein communicating the at least one measured environment related parameter together with the position data of the logger device to the external computer is performed only if the transport vehicle type is determined not to be an aircraft. This is to fulfil the Federal Aviation Administration (FAA) regulations where no transmission of electronic devices having transmitters such as mobile phones and other devices are allowed to be turned on during flight. Thus, increased safety is achieved because it is ensured that no transmission is possible from the take-off until the landing of the aircraft.

At the same time, it is ensured that regular transmission will take place if the logger device is determined not to be onboard an aircraft. During this higher energy mode, the data saved in the memory is transmitted (e.g. an average value of seven data points may be transmitted) to the external computer which provides a real time monitoring of the asses during the transport from the origin location to the destination location. In one embodiment, the step of communicating with the external computer is performed with pre-fixed time intervals or if an excursion is detected by the logger device. This may as an example be where the logger device is set to transmit environment related data together with the location of the logger device to the external computer every hour. Thus, in this embodiment, the transport type would be determined shortly before the transmission to ensure that the transmission is permitted before the schedule, i.e. the logger device is not onboard an aircraft.

In an embodiment, if an excursion is detected the transport vehicle type is determined, where the determined excursion is communicated to the external computer if the transport vehicle type is not determined to be an aircraft. Accordingly, and as already mentioned, it is possible to link excursion rate to transport vehicle types and, in that way, it is possible to optimize the vehicle type selection where the risk of excursion is minimal.

In a second aspect of the invention, a logger device is provided configured to be associated with an asset while the asset is transported from an origin location to a destination location, comprising:

• an accelerometer, where the accelerometer is configured for acquiring acceleration data for the logger device, where the acceleration data comprises a collection of acceleration vectors,

• a processor for controlling the accelerometer by means of receiving the measured acceleration data from the accelerometer and transforming the received collection of acceleration vectors into direction-independent acceleration data, wherein the direction-independent acceleration data is utilized as input data in identifying the transport vehicle type of the vehicle currently transporting the asset.

In an embodiment, the logger device further comprises:

• at least one additional sensor, in addition to the accelerometer, for sensing at least one additional environmental related data,

• a communication module operated by the processor for communicating the at least one measured environmental related parameter together with the position data of the logger device to an external computer by means of temporarily adjusting the communication module from being in a low or off power mode to a higher energy mode, the low power mode being a power mode where communication with the external computer is not possible and the higher energy mode being a power mode allowing transmission of the at least one measured environment related parameter together with position data of the logger device to the external computer,

• wherein operating the communication module by the processor includes only transmitting the at least one measured environment related parameter together with position data of the logger device to the external computer if the identified transport vehicle type is not an aircraft.

In a third aspect of the invention, a system is provided for identifying transport vehicle type of vehicles currently transporting assets, each asset having at least one logger devices associated thereto while the assets are transported from an origin location to a destination location, comprising:

• a computer,

• logger devices each comprising an accelerometer for acquiring acceleration data for the logger device and a communication module,

• a transforming module comprised in the computer or in the logger device configured for transforming the received collection of acceleration vectors into direction-independent acceleration data, wherein the direction-independent acceleration data is utilized as input data in identifying the transport vehicle type of the vehicle currently transporting the asset.

In an embodiment, the computer comprises a storage medium having stored the directionindependent acceleration data that uniquely characterizes the transport vehicle type, wherein the stored direction-independent acceleration data is used as reference data in identifying the transport vehicle type.

In general the various aspects of the invention may be combined and coupled in any way possible within the scope of the invention. These and other aspects, features and/or advantages of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter. BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will be described, by way of example only, with reference to the drawings, in which:

Figure 1 depicts a flowchart of an embodiment of a method according to the present invention,

Figure 2 shows three-dimensional variation acceleration data,

Figure 3 illustrates graphically a method to eliminate direction dependency in acceleration data,

Figures 4-8 show processed direction independent acceleration data for different transport vehicle types, and

Figure 9 shows an embodiment of a logger device according to the present invention configured to utilize processed acceleration data in identifying different transport vehicle types.

DESCRIPTION OF EMBODIMENTS

Figure 1 shows an embodiment of a method according to the present invention of identifying transport vehicle type of a vehicle currently transporting an asset having an associated logger device while the asset is transported from an origin location to a destination location.

In a first step (SI) 101, acceleration data measured by an accelerometer comprised in the logger device is received, where the acceleration data comprises a collection of three- dimensional acceleration vectors.

In a second step (S2) 102, the acceleration vectors are transformed into directionindependent acceleration data so as to eliminate the direction dependency in the acceleration data.

In a third step (S3) 103, the direction-independent acceleration data is used as input data in identifying the transport vehicle type of the vehicle currently transporting the asset.

The transformation into direction-independent acceleration data in step S2 102 may as an example be performed by means of calculating the amplitude variance for the acceleration vectors using the following formula: where a _xi , a _yi , a _zi are the x, y, and z direction components of the three-dimensional vectors i a _z t) and a _x , a _y , a _z are mean values for the x, y and z acceleration components and where n may be defined as a chunk sample number of acceleration vectors collected over pre-defined time-period. The outcome of this calculation is obviously a single directionindependent data value as depicted in figure 3 where the chunk sample is 20 (can of course be several hundred or thousand). This is then repeated multiple times, e.g. several hundred or thousands of times which results in several hundred or thousands data points, which gives the data collection used to identify the transport vehicle type. This will be discussed in more details in relation to figures 4 to 8 showing measurement data where the direction dependency has been eliminated.

The importance of eliminating the direction dependence is illustrated in the scatterplot in figure 2 showing (x,y,z) variance of actual measurements for two logger devices onboard of two aircraft types, Airbus and Boing during take-off. As shown here, the two recording data sets 201, 202 roughly form an ellipsoid. The orientation of the ellipsoid is however different for the two recordings which are caused by the different orientations of the two logger devices. Correcting such differences requires high processing power and thus may greatly affect the battery status of the logger device in case the processing would be performed by a processor comprised in the logger device.

In another example, step S2 may be determined by calculating the Total Variation (TV) of each chunk of n measurements, where for the acceleration along an x-axis: where this is in similar way calculated for y and z-axis. To have a size independent of rotation the values for each axis can be taken together into one size e.g. by taking the Euclidean-norm of the total variation vector tv = (tv _x , tv _y , tv _z ).

Applying the methodology above, and as shown in figure 3 this orientation dependency is eliminated. Figures 4 to 8 depict a histogram with experimental results of an amplitude chunk variance as discussed above and is illustrated in the formula above and in figure 3, for different transport vehicle types. Logger devices with acceleration sensors where into different vehicle types, trains, ships, for days/weeks/months, to obtain solid reference data. This data, which may be seen as a training data for different vehicles, is preferably constantly improved via e.g. Artificial Intelligence (Al), and acts as a fingerprint/reference data for different vehicles in different “modes”, i.e. stationary mode and driving/flying mode.

Figure 4 shows a static variance amplitude for a stationary vehicle which in this case is a truck collected by the accelerometer while the vehicle is standing still, and the engine is running. Similar data would preferably also be collected for other types of vehicles.

Figure 5 shows variance amplitude for the 3D-variance data 202 shown in figure 2 using the principle in figure 3 for one logger device in a first aircraft type, figure 6 shows variance amplitude for the 3D-variance data 201 shown in figure 2 using the principle in figure 3 for the other logger device in a second aircraft type, where the orientation dependency shown in figure 2 has been eliminated.

Figure 7 shows variance amplitude for spiky transit for a vehicle driving in a city (high start/stop frequency rate) and figure 8 shows variance amplitude for a noisy transit for a vehicle driving under stationary conditions such as on the highway.

Moreover, at least one filter may be applied to the direction-independent acceleration data where the resulting data after applying the at least one filter to the direction-independent acceleration data is processed and utilized in identifying the transport vehicle type currently transporting the asset.

Figure 9 depicts graphically an embodiment of a logger device 900 according to the present invention configured to be associated with an asset 915 while the asset is transported from an origin location to a destination location, where the logger device 900 comprises at least one sensor 905 such as temperature sensor, an accelerometer 916 configured for acquiring acceleration data for the logger device, where the acceleration data comprises a collection of acceleration vectors, a memory 901, a power source 902, a communication module 904, and a processor 903 for controlling the at least one sensor, the accelerometer, the communication module and the memory.

In this embodiment, the processor 903 is configured for receiving the measured acceleration data from the accelerometer and transforming the received collection of acceleration vectors into direction-independent acceleration data, wherein the directionindependent acceleration data is utilized as input data in identifying the transport vehicle type currently transporting the asset as discussed in relation to figures 3 to 8.

The step of controlling includes communicating measured environmental related parameter such as the temperature together with position data of the logger device to an external computer 914. The communication module may comprise a modem and where communicating the measured environmental related parameter such as the temperature together with position data of the logger device to the computer may be done by temporarily setting the modem from a low power mode, where no transmission is possible, to a higher energy mode where such a transmission of the data is possible.

The memory is configured to store measured environmental related data such as temperature, acceleration data, humidity, position data of the logger device and light intensity.

The step of controlling the communication module and the memory by the processor 903 includes setting the communication module, e.g. said modem, to a higher energy mode if and only if the transport vehicle type is determined not be an aircraft meaning that the transport vehicle type may be any type of transport, vessel, train etc.. During this higher energy mode the data saved in the memory 901 is transmitted to the external computer which provides a real time monitoring of the asses during the transport from the origin location to the destination location. If however the transport vehicle type is determined to be an aircraft, the means that such a transmission is not allowed according to Federal Aviation Administration (FAA) regulations.

In a preferred embodiment, the processor is scheduled to determine the transport vehicle type each time before a scheduled “wakeup” of the communication module to determine whether it is safe to turn the modem on and send data to our servers. The scheduled wakeup may differ depending on the asset and/or the transport vehicle type, and may as an example be once every hour, or every two hours.

The transport vehicle type may as an example be selected from being any type of vehicle such as a truck 910, 912, or a car, a vessel, train, aircraft 906, 908 etc. wherein each transport vehicle type has a direction-independent acceleration data that uniquely characterizes the transport vehicle type 907, 909, 911, 913 as discussed in relation to figures 4 to 8, i.e. is a kind of a finger print. In the scenario shown here, the result after comparing the measured acceleration data with the reference data may be that the asset 915 together with the logger device 900 is determined to be on aircraft 915 based on the independent acceleration data.

While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.