WO2011116400A1

WO2011116400A1 - Enhanced gps location in mobile asset tracking

Info

Publication number: WO2011116400A1
Application number: PCT/US2011/029284
Authority: WO
Inventors: Herbert Perten
Original assignee: Startrak Systems, Llc
Priority date: 2010-03-19
Filing date: 2011-03-21
Publication date: 2011-09-22
Also published as: AU2011227024B2; NZ602653A; CA2793865A1; CA2793865C; AU2011227024A1

Abstract

Locating a movable asset involves communicating the position of the asset with a GPS on the asset to a network center when the asset is outside a predetermined area, and communicating the position to asset differentially with a GPS on the asset and with a GPS on a reference when the asset is within the predetermined area.

Description

Title

Enhanced GPS Location In Mobile Asset Tracking Related Applications

This application is related to US application 61315590 filed 19 March 2010. Applicant claims priority of this application, and the contents of this applications are incorporated herein as if fully recited herein.

Field of the Invention

This invention relates to automatic asset locations systems, and more particularly to Automatic Vehicle Location (AVL) systems.

Background of the Invention

Automatic Vehicle Location (AVL) systems have gained widespread acceptance in the management of mobile assets such as trucks, trailers and railcars. Typically a satellite based location device, such as a GPS receiver, coupled to a wireless link, such as a cellular telephone, broadcast mobile asset locations to a central station at periodic intervals. The central station presents this information in graphical or tabular form to a fleet manager, dispatcher or other users of this information.

To achieve low hardware cost and low power consumption, AVL systems typically employ single band ('LI') GPS receivers. These give a location in approximately one minute or less and offer an accuracy of approximately 10-12 meters, which is sufficient to determine the location of a moving or stationary asset. However, such simple systems are unable to resolve specific parking spots having spacings about 3.3 meters, loading bays with approximately 6 meter spacings, or parallel rail tracks which are about 3.3 meters apart. More precision location devices are available, but their hardware costs orders of magnitude more than that of the simple GPS devices, and this makes their deployment commercially unviable.

US Patent 7,315,281 to Dejanovic discloses a location system which also does not provide the desired accuracy with low power and low energy.

An object of this invention is overcome the deficiencies of prior asset location systems.

Brief Description Of Embodiments Of The Invention:

An embodiment of the invention uses the GPS position within a specifically defined area to activate a higher accuracy algorithm (differential GPS) than would be used in its normal operation.

An embodiment of the invention involves tracking an asset with a GPS until the asset moves into a defined area and then tracking the asset with a differential GPS.

These and other aspects of embodiments of the invention are pointed out in the claims forming a part of this specification. Other objects and advantages will become evident from the following description when read in light of the accompanying drawings. Brief Description of the Drawings.

Figure 1 illustrates an embodiment of the invention.

Figure 2 shows some of the different ways used to define geofences according to embodiments of the invention.

Figure 3 is a flowchart of the operation of a mobile asset according to embodiments of the invention.

Figure 4 is a flowchart of a network center's response to a message from a mobile asset that it has entered a defined area such as a geofence according to embodiments of the invention.

Figure 5 is a flowchart of the response of a network center to position data received from the mobile asset and from a reference according to embodiments of the invention. Description Of Preferred Embodiments

An embodiment of the invention is shown in Figure 1. Here a mobile asset MAI may be any piece of mobile equipment whose location is to be determined. It may be a trailer, a tractor, a railcar, a locomotive, an automobile, a barge, etc. A GPS receiver GPS 1, mounted on the mobile asset MAI, may be any GPS receiver, and may be in the form of a simple, single band device chosen for low cost and low power consumption. In one embodiment, this is a model LEA-5T made by u-blox (www . u-blox .com) . A motion sensor MS I mounted on or in the mobile asset MAI is, according to various embodiments, a ball balanced on two contacts or a tri-axial accelerometer with a dedicated microprocessor used to detect motion, or any motion detecting device. A two-way radio RAl provides communications. In an embodiment, the radio RAl operates on cellular telephone bands. The radio RAl includes an antenna, and may, according to various embodiments, be a cellular, satellite, Wi-Fi, or any other RF device. The radio RAl communicates over a wireless link WL1. The radio RAl, motion sensor MS I, and the GPS receiver GPS 1 installed on the mobile asset MAI are controlled by a local microprocessor PR1. According to an embodiment, the mobile asset MAI is one of a fleet of mobile devices MAI, MA2, ... MAn each equipped with corresponding sensing GPS, radio and processing equipment as mobile asset MAI.

A reference station RS 1 also contains a GPS receiver GPS2. This may be a simple device like GPS receiver GPS 1 or it may be of a better quality.

Depending on the installation, reference station RS 1 communicates with a network center NCI over wireless or wired lines. If wireless, the radio RA2 and wireless link WL2 may be of the same or different from radio RAland wireless link WL1. A wired connection operates, according to an embodiment, via the internet IC1, though it may use any other suitable network. A processor PR2 controls local operation of the GPS receiver GPS2 and the radio RA2 reference station RS I. One reference station can service multiple mobile assets.

Depending on the quality of its GPS receiver, antenna, and on the accuracy desired, it may offer adequate correction for mobile assets hundreds of kilometers away. According to an embodiment, the reference station RS 1 is one of a number of reference stations RS2 ...RSn. A network center NCI receives data from both mobile asset MAI and reference station RS I. For this purpose it includes one or more radios RA3 to receive wireless signals over wireless links WL1 and WL2 plus an internet connection IC1. A post processor PP1 combines the information received over wireless links WL1 and WL2 from mobile asset MAI and reference station RS 1 to determine an accurate position for mobile asset MAI. In one embodiment, post processor PP1 utilizes post processing firmware in the form of the GrafNav product, supplied by Waypoint Software, a division of Novatel, Inc.

(www.novatel.com) . A single post processor can usually handle all the computations from a multitude of mobile assets and reference stations, though if necessary multiple post processors can share the load.

The processor PRl filters the GPS information before transmitting back to the network center. According to an embodiment, a small processor in the GPS receiver GPS l performs this function instead of the processor PRl. Such filtering involves use of a logical construct known as a "geofence". This is an imaginary boundary ring which may be circular, rectangular or polygonal in shape.

Geofences are drawn around terminals, loading docks or customer sites and messages are sent when assets enter or leave such locations. In effect the processor PRl determines when the GPS location indicates the mobile asset has entered or left a geofence.

Figure 2 shows some of the different ways used to define geofences. 'A' shows a geofence described by a center point and a circle of defined radius. 'B' shows a rectangle defined by the locations of two diagonal corners. 'C shows a polygon defined by its nodes. 'D' is similar to 'B' except that it has been rotated from a north- south-east- west coordinate system. Other embodiments exist without affecting the system. Figure 3 is a flow chart of the logic present on a mobile asset MAI. The decision tree resides in the code of the processor PRl. To save power, the processor PRl keeps the GPS receiver GPS l quiescently off, and only powers it on when a new reading is desired.

The system is initialized by a GPS reading in step 302 where the processor PRl turns of the GPS receiver GPS l, step 304 where the processor PRl takes a GPS reading of the GPS receiver GPS l, and step 306 where the processor PRl turns the GPS receiver GPS l off. This synchronizes a local clock within the processor PRl with the absolute time received from the GPS satellite constellation orbiting the earth. Various timers resident within the code of the processor PRl determine when specific events are to occur. In response to one such timer the processor PRl determines, how often to take a GPS reading. According to an embodiment, the processor PRl actuates the GPS receiver GPS l to take GPS readings every 3 minutes. According to other embodiments the processor PRl actuates the GPS receiver GPS l to take GPS readings at other times such as any one between 30 seconds to 6 minutes or even at different intervals.

In step 308, the processor PRl asks if it is time for a reading. If the answer is no, it is not yet time for a new GPS reading, the processor PRl waits until it is time. If the answer is yes the processor PRl decides whether or not to actually take a new GPS reading. In step 310, the processor PRl interrogates the motion sensor MS I to determine whether the mobile asset MAI is moving or stationary. If the answer is yes, the mobile asset MS 1 is stationary, and there is no need to expend any power to take a new GPS reading, because the position is unchanged from the last GPS reading. Thus in Step 318 the processor PRl retains the previous GPS reading that the processor PRl has memorized. If the answer in step 210 is no, the motion sensor MS I indicates that the mobile asset MAI is in fact moving, then a new GPS reading is required. In response, the processor PRl executes step 312 to turn on the GPS receiver GPS l, executes step

314 to take a GPS reading, and executes step 316 to turn off the GPS receiver GPS l.

Another timer within the processor PRl indicates when to send the GPS location over the radio RA1 and the wireless link WL1 to the network center

NCI. According to various embodiments the processor PRl does this as infrequently as once per day or as often as once per hour. For this purpose, in step 320 the processor PRl asks whether it is time to send a location report. If the answer is no, the processor PRl sends no report, If yes, the processor PRl sends a location report via radio RA1 and wireless link WL1 to the network center

NCI.

The processor PRl stores a listing of known geofences potentially used by the mobile asset MAI. In step 324 the processor PRl asks if the new GPS location determined in steps 314 and 316 falls within one of these geofences. If the answer is no, the processor PRl returns to step 308. If the answer is yes, the processor PRl executes step 326 and sends a message to this effect via radio RA1 and wireless link WL1. The processor PRl then returns to the state indicated at 308 until it is time to get the next GPS reading.

Figure 4 is the logic implemented within the network center NCI as controlled by the processor PR3. Upon receipt of a geofence message from mobile asset MAI in step 326, the processor PR3 starts a timer. In step 402 the processor PR3 asks if the mobile asset has been in the geofence for a time greater than t. If no, then no subsequent geofence message has arrived for a time t and the processor PR3 continues to ask until the time t has elapsed. If the answer is yes the processor PR3 then assumes the asset is still within the geofence. In step 404 the processor PR3 then checks its database to see whether there is a local reference station RS I associated with this geofence. If there is not, the processor PR3 goes back to the head of the loop until such time as it receives a new geofence message from MAI through steps 308 to 324. If the answer is yes, that a reference is associated with this geofence, the processor PR3 starts the procedure for generating a more accurate position determination. In step 406 the processor PR3 sends a message over radio RA3 via wireless link WL2 and radio R2 to the processor PR2 commanding the processor PR2 in reference station RS I to direct the GPS receiver GPS2 to turn on and take GPS readings for tl minutes. In step 408 the processor PR3 sends a message over radio RA3 vial wireless link WL2 and radio RA1 to the processor PR1 commanding the processor PR1 in the mobile asset MAI to direct the GPS receiver GPS 1 to turn on and take GPS readings for tl minutes. In practice, tl is slightly longer than t2, and since it starts first, brackets tl on both ends in time. According to an embodiment, tl is approximately slightly over 10 minutes and tl is approximately slightly less than 10 minutes in length. According to other embodiments, other times sufficiently long to provide accurate locations for particular distances are used. The processor PR1 in mobile asset MAI transmits this data via radio RA1, wireless link WL1, and radio RA3 to the processor PR3 in network center NCI. The processor PR2 transmits its data via radio RA2, wireless link WL2, and RA3 to the processor PR3 in network center NCI. The data collected is managed as shown in Figure 5. The processor PR3 decides how to handle the data. In step 502, when standard GPS signals are received, the processor PR3 passes them to the end user via a post processor PPl in the network center NCI and an internet connection IC1. According to other embodiments the processor PR3 passes the standard GPS signals received directly to the end user without passing them through the post processor PPl.

In step 504 the processor PR3 receives enhanced differential GPS data from the processor PR1 and the processor PR2 in the form of raw data. The processor PR3 then passes the two data streams to the post processor PPl which computes a precision location. In step 506 this information is then passed on to the user in place of the simple GPS information. According to other

embodiments, the internet connection is replaced with a wireless connection.

According to another embodiment, the processor PR3 starts the differential GPS process on command from the user rather than automatically.

According to yet another embodiment, steps 324 and 326 are performed in the processor PR3 of the network center NCI just before step 402. In that case the processor PR3 stores a listing of known geofences potentially used by the mobile asset MAI and any other mobile asset in the system.

According to an embodiment, the post processor PPl determines when to switch from ordinary GPS measurements to differential GPS measurements on the basis of data store that recognizes various geofences on the excursions of the mobile asset MAI. The processor PRl in step 324 in effect constitutes a geofence asset presence detector when the mobile asset MAI is in a geofence. When the mobile asset MAI enters a geofence the processor PRl it in effect constitutes an asset entry detector. The processor PRl in effect constitutes an asset departure detector when the mobile asset MAI departs a geofence. When the mobile asset MAlis outside a geofence, the network center is operating in a non-differential mode or express mode and its connection to the mobile asset forms a non-differential or express coupling via the link WL1.

In battery powered systems, typical of AVL systems, the GPS module's power consumption is a significant portion of the power budget and the module is therefore quiescently kept in an unpowered or very low power "sleep" state. When GPS readings are required, the module is powered up, the reading taken, and the module is once again powered down. Motion sensors allows systems to save the power they would have otherwise spent on activating their GPS systems only to find out they hadn't moved since the last time they checked, i.e. the previous reading was still valid.

At a predetermined interval, the AVL system tries to ascertain its position. GPS readings typically occur at a rate far greater than that which is required for display to the end user. This allows the systems to autonomously determine whether or not they have entered or left geofences. If a particular GPS reading occurs at the time when a location message should be sent, that message is sent at this time. If a geofence has been just entered, that information is transmitted as well.

The flowchart of Figure 4 shows the network center' s response to a message from a mobile asset that it has entered a geofence. Its response to this message is to start a timer. If a time period t has elapsed and no subsequent messages have arrived indicating that the asset is no longer in this geofence, the network center checks its database to see if this particular geofence has a reference station associated with it and whether a precision location is required. According to an embodiment, whether a precision location is required also takes the available power on the asset into consideration, since differential GPS requires data collection for several minutes it imposes significant battery drain. If all the criteria are met, the network center begins the procedure to obtain a differential GPS. According to other embodiments, the network center also uses other information to determine whether or not to obtain a differential GPS, such as bill of lading, time of day, type of geofence, etc.

First, the network center commands the reference station to start collecting raw data. This mode differs from a normal GPS's operation in that the system does not compute its position; rather it just collects the raw data that would normally be used to compute the position. This is done at a rate of once a second for a period exceeding 10 minutes. At the end of the acquisition period, the data is transmitted to the network center. Then the mobile asset is commanded to collect data in a similar fashion. The mobile asset places its GPS system in a special mode wherein it collects and transmits raw satellite data rather than its actual computed position. In an embodiment, the mobile asset is commanded to take readings every second for 10 minutes. The reason the reference station is started first and ends last is to ensure that its data set brackets the mobile asset's data set on both sides in time.

Figure 5 is a flowchart of the response of the network center to position data received from the mobile asset and from the reference. Standard GPS messages receive no further processing and are displayed to the user. Raw data messages are handled differently. Once the completed data sets from both the mobile asset and from the reference have been received by the network center, they are passed to the differential GPS processor or post processor, which computes the accurate position of the mobile asset. In an embodiment using the hardware and software mentioned above, 10 minutes of data acquisition yields better than 1 m resolution. After differential GPS is complete, the system reverts to normal or express GPS.

Transfer from normal GPS to differential GPS and back to normal GPS is important because differential GPS provides greater resolution and getting a GPS reading requires a significant portion of the power budget of a mobile asset. Whereas normal GPS readings may be accomplished in nominally one minute, differential readings require continuous data collection for much longer time periods. In one embodiment, differential GPS data is acquired for 10 minutes. Thus, using differential GPS at all times is unsupportable from a power budget point of view.

In addition, differential GPS requires the uploading of raw data from both the mobile and reference systems. In one embodiment, 10 minutes of data collection results in 17 Mbytes of raw data. The time to upload this may be significant. When this is done wirelessly, there are real monetary costs associated with this action. The power budget is also affected, since the radio must be powered up long enough to transfer this large amount of data.

Finally, with differential post processing of GPS data, the mobile asset does not know its location. This rules out any location-based operation the asset may need to do, such as determining when it has entered or left a geofence. For mobile assets, getting a precise location using differential postprocessing of GPS data as described herein is especially useful when the asset is stationary and when precise location is required. Otherwise, conventional GPS readings are used.

For these reasons, switching between normal GPS and differential GPS offers significant advantages in power use.

According to an embodiment, a mobile asset is designated as a reference for a particular location. This asset must be stationary for a period of time such that its location can be determined using an external reference, such as the CORS, Continuously Operating Reference Stations system operated by the National Oceanographic and Atmospheric Administration (NOAA), whose cumulative correction data is placed on the internet at fixed times, typically 0:00 GMT. In this case, the mobile asset must have been stationary since before 00:00 GMT, since that is when the CORS system is updated. The differential GPS engine computes the actual position of the reference asset using the CORS system. Subsequently, as long as this asset is not moved, it may be used as a reference for other assets in the area.

Embodiment of the invention make accurate tracking possible to use with mobile assets by permitting using differential GPS for mobile assets employing simple GPS systems, using a motion sensor to decide whether to spend the power on a new GPS reading, employing geofences to determine when to activate differential GPS, and using mobile asset as a reference station for other assets.

The invention overcomes the disadvantages of other known location with accuracies in the sub -centimeter range such as that used by the surveying community. In such systems, reference stations with known coordinates are sited, and GPS data from these stations is used to correct the data acquired by the roaming GPS receiver that is moved to points on the survey. This is a form of "differential GPS". In some systems, a wireless link is established between the reference and rover units, and the corrected location is computed in real time by the rover device. In others, the two data sets are merged after completion of the survey. In some cases, the reference stations are sited locally by the surveying teams. The CORS system allows surveyors to correct their field data within a couple of days of taking that data, an acceptable latency for that industry.

Operators of mobile assets cannot wait that long to ascertain the location of their equipment.

An embodiment of the invention uses a network of other reference stations in the public domain. Reference stations may be hundreds of kilometers away and still give adequate correction to local data. The Federal Aviation

Authority (FAA) is in the process of deploying a Wide Area Augmentation System (WAAS) system for realtime GPS correction for aircraft. Reference stations are sited near major airports and the correction data is transmitted to aircraft via a geostationary satellite positioned over the equator. Though planes in the air are always in the line of sight to the equator, ground-based assets on the north sides of buildings (for instance) will be unable to get good correction signals.

If an asset is stationary, averaging of GPS readings may result in increased accuracy, but to get sub-meter accuracy one must acquire data for several hours, which is unacceptably long for users of mobile assets. While embodiments of the invention have been described in detail, it will be evident to those skilled in the art that the invention may be embodied otherwise.

Claims

What is claimed is:

1. A method of locating a movable asset, comprising: conveying position data concerning the movable asset, with an asset GPS on the movable asset, to a network center in a first GPS mode; communicating position data concerning the movable asset to said network center in a differential GPS mode, with the asset GPS on said movable asset and with a reference GPS on a reference station, when said movable asset is within said predetermined area; and passing the location of said asset from said network center a user.

2. A method as in claim 1, wherein: said step of conveying includes storing in said asset a plurality of geofences; detecting when said movable asset has entered a geofence; and switching from the first mode to the differential mode when said asset enters the geofence.

3. A method as in claim 1, wherein the differential mode has a higher resolution than the first mode.

4. A method as in claim 1, wherein: said step of transmitting includes storing in said asset a plurality of geofences; detecting when said movable asset has entered a geofence; switching from the first mode to the differential mode when said asset enters the geofence; and the differential mode has a higher resolution than the first mode.

5. A method as in claim 1, further comprising: switching from the first mode to the differential mode when the asset enters a predetermined area; and switching from the differential mode to the first mode when the asset departs from the predetermined area.

6. A method as in claim 1, wherein: the step of conveying includes detecting position data concerning the movable asset with the asset GPS; and the step of communicating includes detecting differential location data concerning the movable asset relative to a reference station with the asset PGS and with the reference GPS.

7. A method as in claim 1, wherein the step of conveying includes maintaining the asset GPS in a quiescent state and turning the asset GPS on for a given time to take an asset GPS reading.

8. A method as in claim 1, wherein: the step of conveying includes sensing whether the mobile asset is moving or standing still, and maintaining the asset GPS in a quiescent state and turning the asset GPS on for a given time to take an asset GPS reading only when the sensing indicates that the asset is moving.

9. A method as in claim 1, wherein: when said movable asset is within said predetermined area, said step of communication includes said network center commanding said reference GPS to take reference data for a time tl during operation in said differential mode and commanding said asset GPS to take asset data for a time t2 within the time t2 in said first mode.

10. A method as in claim 1, wherein: when said movable asset is within said predetermined area, said step of communication includes said network center commanding said reference GPS to take reference data for a time tl during operation in said differential mode and commanding said asset GPS to take asset data for a time t2 within the time tl in said first mode; and said time tl is approximately slightly more that 10 minutes and the time t2 approximately slightly less than 10 minutes.

11. A movable asset locator system, comprising: a movable asset; a position data GPS receiver on said movable asset; said position data GPS receiver having stored therein a plurality of geofences and a geofence asset presence detector; a reference station having a reference GPS receiver; a data receiving network center operationally coupled to said position data GPS receiver and to said reference GPS receiver in differential mode in response to the geofence asset presence detector, and otherwise having an express coupling to said position data receiver in an express mode.

12. A system as in claim 1, wherein: said network center switches from differential coupling to express coupling in response to the geofence presence detector.

13. A system as in claim 1, wherein the differential mode has a higher resolution than the express mode.

14. A system as in claim 1, wherein said differential mode operates more slowly than the express mode.

15. A system as in claim 1, wherein said network center constitutes an asset departure detector in response to the asset GPS leaving a geofence.

16. A system as in claim 1, wherein said mobile asset includes a GPS initiation timer coupled to the asset GPS to take an asset GPS reading.

17. A system as in claim 1, wherein said mobile asset includes a motion sensor, said mobile asset including a GPS disabler in response to a signal from the motion sensor that the mobile asset is still and a GPS enabler when the mobile asset is in motion.