Classifications

• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
  • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
  • G01S13/87—Combinations of radar systems, e.g. primary radar and secondary radar
  • G01S13/878—Combination of several spaced transmitters or receivers of known location for determining the position of a transponder or a reflector
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
  • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
  • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
  • G01S13/06—Systems determining position data of a target
  • G01S13/08—Systems for measuring distance only
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
  • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
  • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
  • G01S13/06—Systems determining position data of a target
  • G01S13/46—Indirect determination of position data
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
  • G01S15/00—Systems using the reflection or reradiation of acoustic waves, e.g. sonar systems
  • G01S15/87—Combinations of sonar systems
  • G01S15/876—Combination of several spaced transmitters or receivers of known location for determining the position of a transponder or a reflector
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
  • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
  • G01S13/66—Radar-tracking systems; Analogous systems
  • G01S13/72—Radar-tracking systems; Analogous systems for two-dimensional [2D] tracking, e.g. combination of angle and range tracking, track-while-scan radar
  • G01S13/723—Radar-tracking systems; Analogous systems for two-dimensional [2D] tracking, e.g. combination of angle and range tracking, track-while-scan radar by using numerical data
  • G01S13/726—Multiple target tracking
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
  • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
  • G01S13/88—Radar or analogous systems specially adapted for specific applications
  • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
  • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
  • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
  • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
  • G01S13/06—Systems determining position data of a target
  • G01S13/46—Indirect determination of position data
  • G01S2013/466—Indirect determination of position data by Trilateration, i.e. two antennas or two sensors determine separately the distance to a target, whereby with the knowledge of the baseline length, i.e. the distance between the antennas or sensors, the position data of the target is determined
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
  • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
  • G01S13/88—Radar or analogous systems specially adapted for specific applications
  • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
  • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
  • G01S2013/9327—Sensor installation details
  • G01S2013/93275—Sensor installation details in the bumper area

Definitions

• the invention relates to the determination of the locations of multiple objects based on the measurements of multiple sensors. More particularly, the invention relates to determining by trilateration the locations of multiple objects detected by multiple, spaced range sensors .
• Trilateration is the art of determining the location of an object in space based on knowledge of the range (distance) of the object from multiple known locations. For instance, knowledge of the range of an object from a known location (e.g., one particular sensor) defines a sphere on which the object must lie, that sphere being the sphere that is centered at the sensor and has a radius equal to the measured range value.
• the object defines two distinct spheres on which the object must lie. Accordingly, the object must lie on the locus of points defined by the intersection of the two spheres, which is a circle. If the range from a third location (or sensor) to the object is known, then the object is known to lie on the locus of points defined by the intersection of all three spheres. For many practical scenarios, the intersection of these three spheres defines a single point which locates the object .
• range readings from only two sensors to the same object will define two circles that intersect at two points. For many practical scenarios, however, only one of these intersections will be located within the detection area of the sensors.
• a sensor that provides a range measurement, but no bearing measurement is a broad azimuth radar reflection system.
• RF radio frequency
• RF radio frequency
• the radar detection system also provides at least some bearing information.
• Air traffic radar is a well known example of a radar that provides both range and bearing information. Such radars send out very narrow beams from a rotating transmitter antenna. Therefore, range can be determined from the delay of the reflected beam, while bearing can be determined from the angular orientation of the antenna at the time of the receipt of the reflected beam.
• each sensor receives a plurality of reflected wave fronts and, therefore, a plurality of range readings.
• each sensor receives a plurality of reflected wave fronts and, therefore, a plurality of range readings.
• this means that as many as 10 x 10 x 10 10,000 "potential objects" will be identified.
• each of the four sensors has a range reading that defines a circle (or a sphere if a three dimensional system) that intersects a range circle from all three other sensors.
• the invention is a method and apparatus for determining the locations of a plurality of actual objects using trilateration based on the output of a plurality of range sensors.
• a plurality of range measurements are obtained from a plurality of sensors, each sensor capable of providing a multiplicity of range measurements.
• the range measurements from the plurality of sensors are correlated with each other to generate a list of potential objects.
• the list of potential objects is then ordered from highest to lowest likelihood of being an actual object, for example, by ordering the objects according to a calculated cumulative error of the individual sensor measurements upon which the potential object is based.
• the ordered list of potential objects is then pared down to a smaller list of actual objects by selecting the potential object highest on the ordered list and assuming it is an actual object, and then removing from the list all other lower-ordered potential objects that are based on any of the measurements upon which the selected object is based. This process is then repeated for the next highest potential object remaining on the list until all potential objects on the list have either been selected as an actual object or removed from the list.
• Figure 1 is a plan view of an automobile having a sensor system in accordance with the present invention.
• FIG. 2 is a block diagram of an intelligent vehicle control sensor system in accordance with one particular embodiment of the present invention.
• Figure 3 is a range diagram illustrating range measurements of four sensors to a single idealized point object .
• Figure 4 is a range diagram illustrating range measurements in a second environment.
• Figure 5 is a flow diagram illustrating operation in accordance with the one aspect of present invention.
• the present invention pertains to a method and apparatus for sensing range to multiple objects using multiple, spaced sensors, and determining the locations of the multiple objects using trilateration.
• One of the primary problems with such trilateration techniques in complex environments e.g., environments in which there are multiple objects, complex shaped objects, and/or false readings, is paring down the number of potential objects generated by the basic trilateration step of determining intersecting spheres/circles into a map or list of likely actual objects.
• the number of potential objects identified can number in the hundreds, when the number of actual objects is less than a dozen.
• a technique should be implemented to determine which potential objects are likely to be actual objects so that the list of potential objects can be pared down to a reasonably accurate number of actual objects (and their locations) .
• Trilateration techniques can be used in many different applications and environments.
• the invention will be described hereinbelow in connection with several different embodiments relating to automotive use. It should be understood by those of skill in the art that this is not a limitation of the invention and is merely an exemplary embodiment.
• the invention will particularly be described in connection with an embodiment for detecting obstacles in front of a moving car.
• Such a system might be used in connection with an intelligent stop & go system in which the map of obstacles or objects in front of a vehicle is used to control the speed of the vehicle such as, for instance, matching the speed of a vehicle directly in front of the car when operating in a traffic jam environment.
• the discussion thus far has focused on the determination of the location of objects at a given instant in time from the sensor readings.
• the track (i.e., direction and velocity) of the detected objects also is determined.
• a multiplicity of temporally displaced sets of readings are taken, each set used to generate a static map of the locations of the objects in the field of view.
• the multiplicity of maps are correlated with one another in order to determine the tracks (or velocity and direction) of the objects .
• FIG. 1 is a plan view of an exemplary application of the present invention.
• four radar sensors 103 are positioned in a line in the front end of an automobile 101.
• the radar sensors may be pulsed or variable frequency sensors so that they can detect and distinguish multiple reflected wave fronts in order to provide range measurements to multiple objects 111.
• each sensor is a short range 24 GHz pulse radar, such as Model MLAU0003-006 manufactured by M/A- COM of Lowell, Massachusetts.
• each sensor emits a very short radio frequency (RF) pulse and then detects reflections of that pulse back to the sensor.
• the sensor determines the delay between the issuance of the pulse and the reflected wave fronts and calculates the distances to the objects off of which the reflected wave pulses reflected.
• the range of the sensors should be limited to some reasonable distance based on the application. This can be done by discarding any reflected wave fronts below a predetermined magnitude and/or received after a predetermined delay.
• the number of range measurements can be limited to a particular number, e.g., the ten closest objects (reflected wave fronts).
• the number of range measurements can be limited to a particular number, e.g., the ten closest objects (reflected wave fronts).
• the use of more than two sensors helps eliminate . situations where readings from two different sensors can only reduce the location of the potential location of the object to two possible locations rather than one.
• the use of more than two sensors provides the means to reduce false objects in accordance with the present invention when range readings from different sensors of different objects are interpreted as a false object. We have found, that, in the automotive field, four sensors arranged in a line in the bumper of an automobile provide good performance.
• FIG. 2 is a block diagram of an intelligent vehicle control system in accordance with the present invention. It comprises four range sensors 201, 203, 205, and 207 and a digital processor 209 coupled to receive the sensors' range measurement outputs.
• the digital processor 209 takes those measurements, and runs them through an algorithm to create a map of objects in the field of view of the sensors.
• the processor 209 might output that map to a tracking processor 211 that compares the map to previous maps to track the velocity and direction of the objects and generates signals for controlling other components 213 of the automobile, such as the brakes and accelerator, in order to implement the aforementioned stop & go vehicle control system.
• Figure 3 illustrates a simple simulated scenario involving a single object 302 viewed by a four sensor array.
• Boxes 201, 203, 205, and 207 indicate the location of the four sensors, respectively.
• Semicircles 309, 311, 313, and 315 represent the range observation from the four sensors 201, 203, 205, and 207, respectively.
• the center of each range semicircle is the position of the sensor that generated the observation.
• Figure 4 represents a more realistic environment encountered by an automobile.
• each of the four sensors 201, 203, 205, and 207 detect two objects, namely, a person 401 and pole 403.
• each of the four sensors 201, 203, 205, and 207 should detect the two objects and only the two objects, i.e., each sensor should generate two accurate range semicircles.
• one or more of the sensors may (a) not detect one or more of the objects, (b) receive two or more reflections from a single object (particularly if the object is large and/or has a complex shape) , and/or (3) simply receive false telemetry.
• Figure 5 is a simple flow diagram breaking down a trilateration algorithm in accordance with the present invention that generates a list of actual objects from the sensor range measurements.
• Figure 5 basically is a breakdown of the steps that occur in processor 209 of Figure 2.
• the first step 501 in the process is parsing through the range measurements of the plurality of sensors to identify which range semicircles might correspond to the same object in order to create a list of potential objects.
• the first reasonable assumption is that all of the range measurements upon which a potential object is based must be from different sensors.
• circles from different sensors that do not intersect with each other do not correspond to the same object. Note that both of these assumptions are not necessarily true in all cases. However, in the overall scheme, tend not to lead to significant error.
• potential objects only if they correspond to three or four sensor readings, i.e., we exclude any potential objects that are based on only two sensor readings.
• potential objects that are based on only two sensor readings are considered in a distinct algorithm that separately identifies and validates potential objects that are based on only two sensor readings.
• each sensor generating ten readings, there is a total of ten thousand possible potential objects.
• any two range measurements (or semicircles) from two or more sensors that differ by more than the spacing between the sensors will not intersect.
• any one or more of the sensors may detect fewer than ten reflected wave fronts.
• the number of potential objects generated in this first phase is likely to be substantially less than the potential maximum number, but still could number in the hundreds in a real world situation.
• a list of potential objects is generated based on the pertinent range measurements and a location is assigned to each potential object.
• Several algorithms for estimating location based on range measurements that do not intersect perfectly at a single point are known. Any such suitable algorithm may be employed.
• the specific location that is determined for each potential object is a nonlinear problem that can be solved using an iterative least-squares method, such as the Levenberg-Marquardt method. See, for example, William Press et al, "Numerical Recipes in C", Cambridge University Press, 1999. However, we choose a simpler algorithm in order to reduce computational burden.
• the potential object location is calculated using only the range measurement from the two sensors most distant from each other (i.e., the two outermost sensors) of the three or four sensors upon which the potential object is based.
• the possible existence of an object is based on three or four sensor readings, but the assumed location of that potential object is calculated using only two of those range readings, namely the ones from the outermost sensors, and only these two range readings must intersect.
• a Monte-Carlo simulation indicates that the above- described outermost sensor method produces location results the accuracy of which is degraded by less than ten percent compared to the iterative Levenberg- Marquardt algorithm.
• the second phase of the trilateration technique of the present invention is sorting or ordering the list of potential objects according to a metric that provides a reasonable estimate of the likelihood that the potential object corresponds to an actual object (hereinafter termed a "rank metric") .
• a metric that provides a reasonable estimate of the likelihood that the potential object corresponds to an actual object.
• rank metric There are many well-known mathematical algorithms for this purpose.
• a cumulative error for each potential object is calculated.
• the cumulative error is representative of the difference between the range measurements upon which the particular potential object is based and the assigned location of that potential object as determined in the first step 501. More specifically, one simple, yet effective algorithm for generating a cumulative error is a sum of the squares of the errors (or SSE) algorithm.
• the SSE of a potential object that is based on the three readings cannot be compared directly to the SSE of a potential object that is based on the four readings.
• the SSE's can be normalized, for instance, by dividing the SSE by the number of sensor readings upon which the potential object is based.
• the metric is calculated as the sum of the squared errors of the inner sensor range (s) with respect to the location of the potential object.
• Each square error term is further normalized by dividing by sensor-measurement variance.
• the rank metric can be considered to be the sum of one or two random variables, each having zero mean and unity variance. If it is assumed that each variable is also normal and independent, then the ranking metric has a chi-square distribution. Specifically, a three sensor metric has a chi-square distribution with one degree of freedom (from its one inner sensor) , whereas a four sensor metric has two degrees of freedom.
• the rank metric of both three and four sensor potential objects is mapped to the equivalent chi-square value having two degrees of freedom.
• the mapping function can be implemented by a polynomial approximation. Then, the list of potential objects is ordered from lowest metric to highest metric, i.e., from smallest error to largest error. This ordered list essentially is the output of 503.
• Elimination is the third and final step of the trilateration algorithm and is illustrated at 505 in Figure 5.
• the elimination algorithm in accordance with the invention is actually quite simple mathematically.
• the highest ranked potential object on the list e.g., the potential object having the lowest cumulative error
• Every other potential object on the list having a lower rank and which is based on any one or more of the range measurements upon which the selected object is based is assumed to be a false object and is removed from the potential object list.
• the next highest ranking potential object on the list is assumed to correspond to an actual object.
• every other potential object on the list having a lower rank and which is based on any one or more of the range measurements upon which that selected object was based is also removed from the list. The process continues until all objects on the list have been either selected as an actual object or removed from the list.
• the above described trilateration algorithm was applied to the example illustrated in Figure 4, in which an automobile with four sensors 201. 203, 205, 207 is facing a person 401 and a pole 403.
• the algorithm identified two objects as shown in the Figure. Accordingly, the algorithm generated a perfectly accurate map of this particular environment. Specifically, it correctly identified point objects at the location of the pole and the person.
• the pole 403 in this experiment was a three inch schedule-40 PVC pole.
• This pole has a small radar cross-section because of its non-metal material.
• the pole was observed by three sensors 203, 205 and 207, which generated range semicircles 203a, 205a and 207a, but not by the left-most sensor 201. This might have been due to antenna gain considerations.
• the left-most sensor 201 has the least antenna gain toward the pole because its look angle is furthest from bore sight of that particular sensor.
• the person 401 is the closest point object to the sensor array.
• a person does not always present a well-defined reflection point to the sensor array.
• a sensor may detect one leg, the other leg, or the belt buckle. Different sensors may observe different parts of the person.
• the person was observed by all four sensors (see range semicircles 201b, 203b, 205b, and 207b) .
• one sensor 205 reported a slightly short range 205b to the person.
• the trilateration algorithm identified the person 401 as a three-sensor object, and did not associate the fourth sensor range reading 205b as belonging to any object.
• the second from left sensor 203 generated two false observations: one halfway between the person and the pole 203c, and one beyond the pole 203d.
• sensors report false observations. This might occur from electromagnetic interference, multipath interference, or sensor imperfections.

Landscapes

• Engineering & Computer Science (AREA)
• Radar, Positioning & Navigation (AREA)
• Remote Sensing (AREA)
• Computer Networks & Wireless Communication (AREA)
• Physics & Mathematics (AREA)
• General Physics & Mathematics (AREA)
• Radar Systems Or Details Thereof (AREA)
• Position Fixing By Use Of Radio Waves (AREA)
• Testing Or Calibration Of Command Recording Devices (AREA)

Priority Applications (3)

Applications Claiming Priority (2)

Publications (2)

Family

ID=21924350

Family Applications (1)

Country Status (6)

Families Citing this family (33)

Family Cites Families (8)

• 2002
  • 2002-01-09 US US10/042,903 patent/US6522288B1/en not_active Expired - Lifetime
  • 2002-12-17 EP EP02806462A patent/EP1463959B1/en not_active Expired - Lifetime
  • 2002-12-17 CN CNB028279468A patent/CN1320370C/zh not_active Expired - Fee Related
  • 2002-12-17 AU AU2002366996A patent/AU2002366996A1/en not_active Abandoned
  • 2002-12-17 WO PCT/US2002/040347 patent/WO2003060545A2/en not_active Ceased
  • 2002-12-17 JP JP2003560587A patent/JP4145800B2/ja not_active Expired - Fee Related

Also Published As

Similar Documents

Legal Events