• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The invention relates to a method for locating an animal by means of radio waves. When radiolocation arrives at a number of possible calculation results for the position of the animal. Stochastic calculations are used to filter out from the majority of the current position calculation results the result which has the highest probability, based on the results of previous measurements and calculations with respect to the possible positions of the node to be located. The stochastic calculations include acceleration values measured on the animal. It is assumed that increasing the measured acceleration values increases the likelihood that even larger distances between temporally successive positions of the node to be located can occur.
    公开号:AT517225A4
    申请号:T374/2015
    申请日:2015-06-12
    公开日:2016-12-15
    发明作者:Wolfgang Auer；Branislav Rudic；Markus Pichler
    申请人:Smartbow Gmbh；
    IPC主号:
    专利说明:

    description
    The invention relates to a method for locating animals by means of radio waves. Conventional installations for the location of objects, such as animals, by means of radio waves, have a plurality of radio transmitters and / or radio receivers located at known positions, and in each case a radio transmitter and / or radio receiver at each object to be located.
    Furthermore, the transmitters or receivers which are used for the detection by means of radio waves are referred to simply as "nodes". In accordance with a frequently used method, in order to determine the position of the node attached to the animal to be located, in the first step radio signals are used to measure the length by which the distances of the individual nodes of known position to the node to be located differ from one another. For example, from the node to be located, a signal is simultaneously sent to all other nodes. The receiving nodes measure the time at which this signal arrives at them. The differences between the individual measured times, in each case multiplied by the speed of light (signal propagation speed in the relevant medium), give the distance differences of the individual nodes to the transmitting node. For the further calculation, it is assumed in the first step that the node to be located is located on a hyperboloid whose axis passes through two knots of known position as foci, the measured distance difference of these nodes to the node to be located being equal to that length which, by definition, differ in the distances between the two foci to each point on the hyperboloid. By cutting at least three such hyperboloids, the possible position of the device to be located is limited to two points. The
    Moi Ί-οτο Vi nQ ^ hannleiinn snf di ηαπ point "can be done mi 1" Hi 1 fp of the third hyperboloid (so that at least four knots of known position are required) or by using known geometric conditions a point anyway can be excluded, for example, because he is outside of the stable within which animals can move, is located. (By "hyperboloid" in this text is meant a rotationally symmetric cup-shaped surface which may be thought of as being due to rotation of a hyperbola about its major axis.)
    If in a modification to the described method in the first step by radio waves directly on the distances between the node at the animal to be located to the individual nodes known position is closed, so in the second step instead of the hyperboloid shells to accept spherical shells.
    With radio localization of animals according to the principle explained, for example, the documents AT 506628 Al, US 6122960 A, DE 100 45 469 C2, W09941723 Al WO2011153571 A2 and W02012079107 A2 are concerned.
    Due to considerable measurement errors and measurement inaccuracies, which are often unavoidable in practice - e.g. due to reflections from radio waves - additional logical assumptions have to be made and corresponding evaluations made in order to obtain a reasonably reliable locating result. In addition to the above-mentioned exclusion of results that are impossible due to geometric conditions, especially stochastic methods are used in order to limit the ambiguity of the current results by using the results of previous measurements and to find the measurement result which has the lowest probability of error Reality reflects. In this context, a proven stochastic model is the Hidden Markov Model and, in particular, the Viterbi algorithm, with the aid of which the currently most probable state sequence can be found relatively efficiently from a multiplicity of possible state sequences.
    A boundary condition usually used for the calculation of the probability of residence is that, at least over a certain limit distance between a location currently being examined and the last accepted location, the probability decreases with increasing distance that the location currently being examined is the current location. Expressed simply, this means that a current location of an animal with high probability can only lie within a circle with a certain, limited radius from the last previously assumed location.
    For example, EP 1 494 397 A2 describes a method of radio localization, in particular for use in buildings. In buildings, radio localization is particularly difficult due to frequent signal reflections.
    For example, EP 549081 A1, GB 2234070 A, GB 2278198 A, US 3999611 A, US 6122960 A, US 7616124 B2, WO 2002091001 A1, WO 2003055388 A2, WO 2006077589 A2, WO 2010108496 A1 and WO 2010109313 A1 propose and explains how to attach acceleration sensors (among others) to living animals and to deduce the behavior of the animals from the measurement results of the acceleration sensors which leads to the respective accelerations. Usually, the measurement results are transmitted via a radio link to a data processing system and checked by this for matches with stored as a pattern of temporal progression of acceleration data.
    The temporal patterns stored as a pattern are characteristic of certain activities of the animal such as walking, eating, rumination, sleeping, possibly limping, riding on other animals. In order to find the characteristic patterns, in earlier work acceleration data and at the same time the activities of animals determined by direct observation were recorded and correlations between acceleration patterns and activities were filtered out from the recorded data.
    W09941723 Al deals with a device carried by a human or animal which can transmit and receive radio waves and whose position can be detected by a satellite navigation system. It is also mentioned that the device can also have an acceleration sensor in addition to various other sensors that can measure a biological state, for example.
    WO2011153571 A2 and WO2012079107 A2 are concerned with radio-enabled ear tags for animals, wherein an ear tag both enables radiolocation and can contain an acceleration sensor, with the aid of which animal activities can be automatically detected by pattern evaluation.
    The US 6122960 A is mainly concerned with the measurement and recording of movements and distances traveled by humans or animals by measuring accelerations and evaluation of the measurements (two-fold integration of the measured acceleration vectors over time). It is proposed in addition to determine an "absolute position" by radio navigation.
    Starting from this prior art, the object of the invention is to provide an automatically executable method for the detection of animals by means of radio waves also applicable in pens and paddocks for animals, which compared to known such methods relative to the investment required for this investment more accurate and provides better reliable results.
    To solve the problem, the known method of radiolocation is used, according to which stochastic calculations are used to filter out the result of previous measurements and calculations with respect to the possible position of the node to be located, from the majority of the respective current position calculation results, which actually describes the current position with the lowest probability of error, and includes as a boundary condition that at least over a certain limiting distance between the position defined by a current calculation result and the last assumed position of the node to be located, with increasing distance between the two positions, the probability decreases that the position according to the current calculation result is the actual position of the node to be located.
    According to the invention, it is proposed to improve the method by also measuring acceleration data on the animal to be located and including the measured acceleration data in the probability such that when higher acceleration values are measured, the probability of greater distance between two temporally successive occupied locations to the detriment of Probability of smaller distance between two temporally successively occupied whereabouts is assumed to be increased.
    The invention will be clarified with the aid of drawing to an advantageous exemplary variant of the method according to the invention:
    Fig. 1: illustrates very stylized, the task from many calculated whereabouts find out the sequence of whereabouts that corresponds with the highest probability of reality.
    2 shows in Cartesian coordinate representation two assumed probability profiles a, b for the size of
    Distances between possible positions (whereabouts) of the node to be located which are found chronologically successively through measurement and calculation.
    A radio location system for determining the location and movements of an animal has - as explained in more detail above - several nodes, wherein the animal, which can move freely carries the node to be located and a plurality of other nodes are immovably mounted and their Relative position is known to each other. At regular intervals, a measurement and calculation procedure for the radio location is performed. As described above, the measurement is either a distance measurement between the node to be located and the nodes of known position or the measurement of the differences in the distances between the node to be located and the individual nodes of known position. From the measurement results, geometric calculation is used to deduce the possible position of the node to be located (relative to the nodes of known position). Due to the fact that very often more than four nodes are known position and that radio signals between two nodes are normally transmitted not only in a direct straight line, but also reflection over additional, longer paths, a multitude of measuring events falls to the one mathematically as an over-determined, self-contradictory system. Specifically, this means that, if one always uses the last four reception events of radio signals between the node to be located and another known position node for calculating the position of the node to be located, one obtains a plurality of calculation results, each one position in space, but only a small portion of these computational results actually describe the positions of the node to be located. The vast majority of the calculation results, however, is wrong.
    In Fig. 1, this relationship is symbolized. A plurality of calculation results each denoting a position c (each symbolized by an oblique square) are followed by a plurality of calculation results each meaning a position d (each symbolized by a star), etc., about calculation results on positions e and positions f ,
    From the outset one can assume as a fixed fact that on the actual path of the node to be located to a single one of the positions c, a single one follows the positions d, then a single one of the positions e and finally a single one of the positions f. Representing the myriad of possible ways in principle possible in Fig. 1, for example, a single path g symbolized by a dotted line.
    To calculate which way is most likely to reflect reality, consider that the possible distances between two consecutive positions (c and d, d and e, e and f) are not all equally probable, but some Classes of distances are more likely and others less. Extremely long distances, for example, are impossible because they would mean that the animal carrying the node to be located would have to move faster than the limits of biology and technology allow.
    FIG. 2 shows a diagram which contains statements regarding the probability of amounts of distances between two positions directly one behind the other.
    The two curves a, b respectively describe how much a distance Dl, D2 between two temporally successive positions is likely.
    If z. B. curve a applies and for a time exactly two possible distances Dl and D2 are calculated, so shows the size ratio of the distances Dl and D2 associated ordinate measures
    Pal and Pa2 to each other the ratio of the probability that Dl applies to the probability that D2 applies. For the further calculation, the absolute size of the probabilities must be normalized so that the sum of the probabilities over all possible positions always yields the same value (preferably 1).
    Both curves a and b have the maximum at D = 0, which according to both curves distances between temporally successive positions are the more probable, the smaller the distances are.
    However, curve a drops significantly steeper than curve b. That is, if curve a is for assuming the probability, the probability that the larger distance D2 is the true is much less than the probability that the smaller distance Dl is the better, than if curve b is the probability of the probability serves.
    According to the invention, the selection of which curve a or b is used depends on which accelerations are measured on the acceleration sensor carried by the animal. If small acceleration values were measured during the time period for which the position is to be calculated, the steeper curve a is used, and if larger acceleration values were measured, curve b is used.
    For example, the curves a, b may be assumed to be Gaussian normal distributions, and the assumed standard deviation, which is known to be involved in the calculation formula of the curves, may be assumed to be dependent on the measured acceleration. The dependency function should preferably be such that the standard deviation increases monotonously with increasing acceleration, for example, increases in direct proportion to the measured acceleration.
    Optimal curves and dependencies as the shape of the curves depend on the measured accelerations must be determined empirically. For this purpose, movements of animals - or robots, etc. - are logged on the one hand by direct observation, distance measurement and recording and on the other hand as described by radio location, acceleration measurements and merging calculation determined. The merging calculation is varied with different underlying calculation parameters (for example, different dependencies of the standard deviations of curves according to Fig. 1 from the measured acceleration) until the two determination methods provide the best possible equally good results.
    If all the calculation parameters have been correctly assumed and are included in the calculation, the best of the possible paths g according to FIG. 1 can be found by forming the respective products of the three probabilities corresponding to the respective distances cd, de from all possible paths along position sequences cdef and ef are assigned as shown in Fig. 1, and then the path at which this product is highest is selected. (In order to keep the computational effort within tolerable limits, the well-known methods Hid-the-Markov-Model, in particular Viterbi algorithm, mentioned in the introduction should be used.)
    Specifically, the acceleration value arriving in the calculation of the probabilities is to be understood as a mathematically processed numerical value which is formed from the majority of the acceleration values measured in the respective current time period and represents these plurality of values well. It may be a statistical mean, such as the root mean square (RMS) value, or the geometric sum of the rms values measured in single directions, or the arithmetic mean, or the average of the individual absolute values, etc., but it may also be a weighted mixture of various such averages or the magnitude of a vectorial sum of the individual detected acceleration vectors. In practice, the method of determination in question is best determined empirically. The selection depends not only on the theoretical mathematical precision but also, for example, on the measurement frequency, the measurement accuracy, the available computing capacity, etc.
    As a result of monitoring the movements of an animal, it is particularly interesting to know how much way the animal has traveled in a period of time (for example in one day), because this tells a great deal about the degree of activity of the animal and thus about the condition of the animal. The knowledge of the exact positions that the animal held at each time point, however, is less meaningful and thus less interesting.
    Especially when it comes to monitoring the entire journey that an animal takes in a "longer period of time" (eg an hour or a day), errors of the radio positioning relatively speaking affect very strongly, if the animal makes little movement and thus even low accelerations can be measured.
    Therefore, according to a preferred further development of the method according to the invention, it is recommended to include in the calculation of the most probable trajectory traveled by an animal-mounted node of a radio location system, results of the radio location system obtained while high acceleration values were measured are more likely to be true than results of the radiolocation system that were obtained while measuring low acceleration values.
    There are a variety of ways to bring this into the concrete calculation of the most likely to be considered trajectory. As an extremely simple example method, it is possible to simply ignore measuring results of the radio positioning, which apply to time ranges in which the magnitude of the measured acceleration was below a certain minimum limit. The finally assumed trajectory then proceeds in the most direct possible way only between those possible result positions of the radiolocation which apply to times at which accelerations were measured which are above the said minimum value.
    权利要求:
    Claims (4)
    [1]
    claims
    A method for locating an animal by means of radio waves, wherein the animal is equipped with a location to be located node of the radio location system and wherein a plurality of further nodes of the radio location system are arranged in known positions, wherein radio signals between the node to be located and the nodes with known Position are transmitted and measured from the signal transmissions parameters individual distances and / or individual distance differences between the individual nodes known position on the one hand and the node to be located on the other hand are calculated and calculated by a data processing system of several such calculation results possible positions of the node to be located, stochastic computations are used to determine the results of previous measurements and computations with respect to the possible position of the node to be located, of the majority of the respective current position calculation results Filter out the result that actually describes the current position with the lowest probability of error, with at least a certain limit distance between the position defined by a current calculation result and the last assumed position of the node to be located, with increasing distance between the two Positions that reduce the likelihood that the position according to the current calculation result is the actual position of the node to be located, wherein also acceleration data are measured at the animal equipped with the node to be located and the measured acceleration data are transmitted to the data processing system, characterized in that the assumption of said probability depends on the measured acceleration data, with larger measured acceleration values the probability of greater distance between two consecutive times taken positions to the detriment of the likelihood of smaller distance between two successive positions, is assumed to be increased.
    [2]
    2. The method according to claim 1, characterized in that the probability for the size of a distance between two temporally successive positions of the node to be located as a function of the amount of the distance as at least approximately a Gaussian normal distribution is assumed following, the standard deviation as with the measured acceleration is assumed to increase monotonically.
    [3]
    3. The method according to claim 1 or 2, characterized in that is used as the in the assumption of probability flowing in acceleration value, a statistical mean of temporally successive measured values of accelerations.
    [4]
    4. Method according to one of claims 1 to 3, characterized in that, in the calculation of the most probable sequence of a series of positions assumed by the node to be located, the possible positions of the node to be located resulting from the results of the radio positioning are more probable than If the acceleration measurement is simultaneous with the radiolocation, higher acceleration values are detected than if lower acceleration values are detected during acceleration measurement at the same time as the radiolocation.
    类似技术:
    公开号 | 公开日 | 专利标题
    EP3104694B1|2020-12-02|Method for locating animals
    DE60207633T2|2006-08-17|Method for eliminating the false reflected targets and for automatic reflector mapping in secondary surveillance radar
    DE102015214338A1|2017-02-02|Determining an arrangement information for a vehicle
    DE102004052110B4|2018-08-23|Method for level measurement according to the transit time principle
    DE112007003538T5|2010-05-20|Position tracking device and method
    DE102018104090A1|2018-09-06|ADAPTIVE DESCRIPTION OF PROCESS RUSTS FOR IMPROVED KALMAN FILTER TARGET TRACKING
    DE19633884A1|1998-03-05|Procedure for fault-tolerant position determination of an object
    DE102016222272B4|2018-05-30|Appreciating an own position
    DE3835992A1|1989-05-03|COLLISION PREVENTION SYSTEM
    EP3308191B1|2019-07-24|Method for locating animals using radio waves
    DE102014003152A1|2015-09-03|Method and device for determining at least one object parameter of a means of transport
    EP2427781B1|2013-06-26|Design for determining an estimated value for a position of a reception element
    DE3740656A1|1989-06-15|NAVIGATION METHOD AND DEVICE FOR DETERMINING THE RELATIVE POSITION AND CALCULATING THE ARTIFICIAL AZIMUT
    EP1352375B1|2005-08-24|Method and device for estimating movement parameters of targets
    EP3295126B1|2019-05-08|Method for determining states of a system using an estimation filter
    WO2011036013A1|2011-03-31|Method and device for determining the position of a watercraft
    EP2686640A1|2014-01-22|Method and device for fusing partitioned correlated signals
    EP2699933B1|2016-11-23|Method and device for determining target parameters
    EP2573583B1|2017-04-26|Detector and method for detecting a transmitter
    DE10242808B4|2018-09-06|Method and device for detecting obstacles and / or road users
    DE10153212A1|2003-06-05|Navigation system for determining the course of a vehicle
    DE102020107950A1|2021-09-23|Procedure for fault detection and safe sensor system
    EP3532317B1|2021-05-19|Method for operating a tyre pressure monitoring unit and tyre pressure monitoring system
    DE102017220506A1|2019-05-16|Method and device for estimating a maritime traffic condition
    DE10238896B4|2006-12-28|Method for evaluating radar data
    同族专利:
    公开号 | 公开日
    CN107690588B|2021-08-24|
    WO2016197160A1|2016-12-15|
    RU2017144307A3|2019-09-19|
    PL3308191T3|2020-01-31|
    AU2016275540A1|2017-12-14|
    AU2016275540B2|2020-04-09|
    US11035924B2|2021-06-15|
    RU2017144307A|2019-07-12|
    EP3308191A1|2018-04-18|
    AT517225B1|2016-12-15|
    BR112017026541A2|2018-08-14|
    CN107690588A|2018-02-13|
    US20180160650A1|2018-06-14|
    RU2702756C2|2019-10-11|
    EP3308191B1|2019-07-24|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US3999611A|1974-11-18|1976-12-28|Louis Bucalo|Devices relating to hooves|
    GB2234070A|1989-06-12|1991-01-23|Anthony Alexander Scarisbrick|Analysis of performance during an exercise activity|
    NL9102182A|1991-12-24|1993-07-16|Stichting Inst Mech|METHOD AND APPARATUS FOR DETERMINING THE STATE OF AN ANIMAL|
    GB2278198B|1993-05-20|1997-01-29|Mini Agriculture & Fisheries|Condition indicator|
    US6122960A|1995-12-12|2000-09-26|Acceleron Technologies, Llc.|System and method for measuring movement of objects|
    US6317049B1|1998-02-17|2001-11-13|Souhail Toubia|Apparatus and method for locating missing persons, animals, and objects|
    US6113539A|1999-01-27|2000-09-05|K.E.R. Associates, Inc.|Physical monitoring system for feedlot animals|
    EP1109031A1|1999-12-10|2001-06-20|Ascom Systec AG|Method and apparatus to localise a mobile terminal|
    DE10045469C2|2000-09-14|2002-12-05|Reinhard Ludwig|Electronic monitoring of a herd of animals|
    KR20010074568A|2001-05-08|2001-08-04|이중호|The acceleration sensing motion sensor and Method for sensing therof|
    DE10164534A1|2001-12-31|2003-07-10|Dirk Parchmann|Device and method for determining parameters of the movement of a body|
    US7250907B2|2003-06-30|2007-07-31|Microsoft Corporation|System and methods for determining the location dynamics of a portable computing device|
    ITBA20040059A1|2004-12-23|2005-03-23|Matrix Srl|LOCALIZATION SYSTEM FOR PEOPLE, ANIMALS AND THINGS, BY INNOVATIVE NETWORK OF TRANSCEIVERS WITHOUT CABLES AND LOW ENERGY CONSUMPTION|
    IL166394D0|2005-01-19|2006-01-15|Vladimir Voronin|A system and apparatus for detecting estrus|
    AR053485A1|2005-05-24|2007-05-09|Commw Scient Ind Res Org|A SYSTEM TO ELECTRONICALLY CONTROL AND MONITOR ANIMALS|
    US7616124B2|2005-10-11|2009-11-10|Snif Labs, Inc.|Tag system|
    US7904097B2|2005-12-07|2011-03-08|Ekahau Oy|Location determination techniques|
    US7592909B2|2006-01-19|2009-09-22|Board Of Regents, The University Of Texas System|Location and tracking system using wireless technology|
    US8055469B2|2006-03-03|2011-11-08|Garmin Switzerland Gmbh|Method and apparatus for determining the attachment position of a motion sensing apparatus|
    KR100815152B1|2006-11-07|2008-03-19|한국전자통신연구원|Apparatus and method for integrated navigation using multi filter fusion|
    US7821406B2|2007-02-27|2010-10-26|Wangrud Carole A|System for monitoring and tracking animals|
    AT506628B1|2008-03-27|2012-01-15|Schauer Maschinenfabrik Gmbh|DEVICE FOR MONITORING ANIMAL ESTABLISHMENTS HELD IN A STABLE|
    US20110102154A1|2008-05-05|2011-05-05|Smarter Farming A/S|Detection of moving objects|
    US20100130230A1|2008-11-21|2010-05-27|Qualcomm Incorporated|Beacon sectoring for position determination|
    WO2010108496A1|2009-03-25|2010-09-30|Smarter Farming A/S|System and method for detecting behaviour of animals|
    ITPV20090004A1|2009-03-27|2010-09-28|Promogreen Com Srl|SYSTEM FOR LOCALIZATION AND TRAINING|
    RU2432581C1|2010-03-03|2011-10-27|Общество с ограниченной ответственностью "РТЛ-Сервис"|Method to locate radio centre, system of radio centre location and unit of data processing|
    AT509892B1|2010-06-09|2014-06-15|Mkw Electronics Gmbh|ANIMAL BRAND|
    AT13366U1|2010-12-15|2013-11-15|Mkw Electronics Gmbh|A method of displaying information associated with an animal|
    DE102011000250A1|2011-01-21|2012-07-26|Vorwerk & Co. Interholding Gmbh|Method for determining the position of a self-moving device|
    WO2012167301A1|2011-06-10|2012-12-13|Navisens Pty Ltd|Positioning, tracking and trajectory estimation of a mobile object|
    US8706137B2|2011-08-02|2014-04-22|Qualcomm Incorporated|Likelihood of mobile device portal transition|
    EP2615598B1|2012-01-11|2017-12-06|Honda Research Institute Europe GmbH|Vehicle with computing means for monitoring and predicting traffic participant objects|
    WO2014038041A1|2012-09-06|2014-03-13|株式会社東芝|Position detection device, position detection method and position detection program|
    CN103134494B|2012-12-05|2015-07-29|中国科学院深圳先进技术研究院|Automobile navigation method in indoor environment and device|
    US8934921B2|2012-12-14|2015-01-13|Apple Inc.|Location determination using fingerprint data|
    AT515088B1|2014-02-14|2015-06-15|Mkw Electronics Gmbh|Method for locating animals using radio waves|
    AT515870A2|2014-06-05|2015-12-15|Mkw Electronics Gmbh|Data network for the surveillance of animals|
    GB201500411D0|2014-09-15|2015-02-25|Isis Innovation|Determining the position of a mobile device in a geographical area|
    EP3585152A4|2017-02-27|2020-03-04|Agersens Pty Ltd|Wearable apparatus for an animal|NL2025037B1|2020-03-03|2021-10-14|Nedap N V|Location validation unit for validating a location of at least one animal, livestock management system, method and computer program.|
    法律状态:
    优先权:
    申请号 | 申请日 | 专利标题
    ATA374/2015A|AT517225B1|2015-06-12|2015-06-12|Method for locating animals using radio waves|ATA374/2015A| AT517225B1|2015-06-12|2015-06-12|Method for locating animals using radio waves|
    RU2017144307A| RU2702756C2|2015-06-12|2016-06-09|Method of determining animal location using radio waves|
    PL16739010T| PL3308191T3|2015-06-12|2016-06-09|Method for locating animals using radio waves|
    PCT/AT2016/000064| WO2016197160A1|2015-06-12|2016-06-09|Method for locating animals using radio waves|
    CN201680033017.4A| CN107690588B|2015-06-12|2016-06-09|Method for locating animals using radio waves|
    AU2016275540A| AU2016275540B2|2015-06-12|2016-06-09|Method for locating animals using radio waves|
    EP16739010.3A| EP3308191B1|2015-06-12|2016-06-09|Method for locating animals using radio waves|
    BR112017026541-9A| BR112017026541A2|2015-06-12|2016-06-09|process for locating animals with the aid of radio waves|
    US15/580,305| US11035924B2|2015-06-12|2016-06-09|Method for locating animals using radio waves|
    [返回顶部]