WO2000002013A1

WO2000002013A1 - Position measurement system

Info

Publication number: WO2000002013A1
Application number: PCT/DK1999/000370
Authority: WO
Inventors: Mogens Blanke; Troels Winther Udsen
Original assignee: 3D Sensor Aps
Priority date: 1998-06-30
Filing date: 1999-06-30
Publication date: 2000-01-13
Also published as: AU4768499A; DK199800852A; EP1092130A1; DK174336B1; US6469501B1

Abstract

Position measurement system for measurement of the relative position between two objects. The system comprises a first and a second inductive system. The first inductive system comprises a transmitting coil and a first detector, with the attribute that the pattern of conductors that form the coil extends along a longitudinal axis. A second inductive system comprises a second transmitting coil and a second detector, with the attribute that the coil extends essentially in the longitudinal direction. The second transmitter coil can be realised as a Meander pattern of conductive material, with return path in its center-line. Using said detectors, it is possible to accurately determine the linear and/or angular position(s) of two bodies relative to each other.

Description

Position measurement system

This invention is concerned with a system for the measurement of linear and angular positions of one object relative to another, comprising a first inductive system that consists of a first transmitting coil and a first detector; a second inductive system that consists of a second coil and a second detector, wherein the said first transmitting coil has conductive elements arranged in a geometry such that displacement along one axis will change the magnetic flux through a selected area in a predetermined way, whereas the flux through the said area will remain essentially unchanged when the displacement is perpendicular to the said axis. Two or more sets of transmitter coils and detectors are used to sense displacement in several dimensions.

An electromagnetic system for the measurement of relative positions is a transducer that will convert a relative physical position and angular rotation between two bodies to a signal that can be interpreted as their relative position and rotation. Transducers for linear and angular position are applied in a multitude of technical and mechanical devices. In the following this will be referred to as a position measurement system.

A position measurement system is concerned with two objects which are movable relative to each other and for which the relative linear and or angular positions shall be determined. Each of the two objects will have components that can generate and/or detect an electromagnetic field. By generating an electromagnetic field at one object that has an predetermined spatial variation, detection at the other object will make it possible to record pairs of related values of electromagnetic flux and the relative position of the two objects. Detection of the magnetic flux thus makes it possible to determine said position between the two objects. The accuracy with which this can be accomplished depends on how well this relation can be established. As a general observation, the transmitted field could be generated in close proximity to the place of detection to avoid disturbances from other electromagnetic fields, including the Geomagnetic field.

Applications of position measurement systems include measurement of rotation about one or more axes and/or measurement of linear displacement in one or more dimensions. Some position measurement systems are generally applicable, others are dedicated to the measurement of eg rotation about one axis or linear displacement in one dimension.

In connection with measurement of more advanced patterns of relative motion, ie relative position and/or rotation in more than one degree of freedom, a trade-off must usually be made where accuracy is significantly reduced to obtain measurement in more than one degree of freedom. This is caused by mechanical limitations to the layout of transmitter and detector elements for measurement in two or three dimensional structures. The result is that a high accuracy can usually not be achieved.

US patent No 5 523 683 describes an induction sensor to measure the displacement of one body relative to another using electromagnetic induction. The paper describes a sealer and a slider, each made as hollow cylinders on each of which a coil is constructed as a set of conductors. The induction sensor measures linear displacement along said cylinder, with the coils generating a field that is essentially varying along the axis of the cylinder, with the exception of a stray component from feed and return conductors axially on the cylinder. The trans^¬ mitter and receiver coils are made as geometrically equivalent in a pattern to make it insensitive to any other motion than the desired linear displacement.

This induction sensor is thus capable of measuring linear displacement only and is insensitive to sealer and slider being rotated at an angle relative to each other.

GB patent application No 2 095 840 describes a system for the determination of position information based on magnetic induction. This system comprises a permanent magnet and magneto-sensitive semiconductor devices located around the permanent magnet.

This system has the drawback that the magnetic field is constant over time and thus rather sensitive to other magnetic fields, including the geomagnetic field and fields from magnetised objects. The geometry of this sys- tern is not made to compensate for such disturbances.

GB patent application Nom 2 197 078 describes a system for the determination of relative positions comprising a transmitter and a receiver. The transmitter comprises three coils arranged along orthogonal axes. The coils are driven pairwise by an alternating current, resulting in a field that rotates about the axis of the second coil. This field is detected using the receiver which can be a coil. The induced voltage in the coil can then be related to the position of the receiver relative to the transmitter. The receiver can comprise three orthogonal coils.

This system is not always useful since the achievable accuracy depends on the distribution of conductive material in proximity to receiver and transmitter. Furthermore, the geometry of the electromagnetic field is relatively complex which hinders a very accurate determination of position.

It is thus a difficulty with known electromagnetic posi- tion measurement systems to determine the position of an object in more than one degree of freedom in that the mechanical construction causes the generated electromagnetic field to have a complexity that in practice hinders the detection of the field such that the detected signals can be used for precise determination of relative position. It is a further obstacle that existing solutions do not make it possible to determine the achievable accuracy through simple geometry of the mechanical construction of the sensing device.

One purpose of the present invention is therefore to provide a position measurement system which can establish an electromagnetic field that can be precisely detected and uniformly in at least two degrees of freedom and with a desired accuracy.

This is achieved by constructing the aforementioned position measurement system such that the second transmitting coil comprises electrical conductors arranged in an es- sentially parallel way, which extend in a direction perpendicular to the axis in which a displacement is desired to be sensed, and distributed along the axis such that a second electromagnetic component can be established perpendicular to the said axis. The conductive elements are arranged in a geometry such that displacement along the said axis will change the magnetic flux through a selected area in a predetermined way, whereas the flux through said area will remain essentially unchanged when the displacement is perpendicular to the axis. Two or more sets of transmitter coils and detectors are used to sense displacement in two or three dimensions. Conductor elements can be distributed along the said axis and a desired accuracy can be obtained using commonplace design. The result is a position measurement system that can establish a well-defined electromagnetic field, that can be easily detected and has a spatial variation along two or more orthogonal axes.

In a preferred embodiment the first transmitter coil com- prises a second conductor element which is essentially parallel with the first conductor element. This enables an increase of the transmitted field strength.

In a specifically preferred embodiment, the first and second conductor elements have sufficient separation for current through the conductors generates (to generate) an electromagnetic field that can be detected with the first detector. This makes it possible to generate a well defined, spatially varying electromagnetic field in a de- sired region around the object.

A very compact geometry can be obtained when the first transmitting coil is ring-shaped.

In a particular form, the conductor elements in the second transmitter coil are connected to make a conductive pattern around a mid-line. The pattern can then be made according to a desired geometry of the detector such that a position dependent signal can be generated with desired properties.

In a particular form, the second transmitter coil is given a Meander pattern with the mid-line as axis of symmetry. A Meander coil has a periodically repeated pattern that can be repeated along its longitudinal axis and thus give a uniform longitudinal detectability along the en- tire coil. The return current is led through a conductor at the said mid-line, thus shaping the electromagnetic components as desired along as well as perpendicular to the longitudinal direction.

In another form, the first detector has a geometry that overlaps the first transmitting coil perpendicular to the longitudinal direction, such that a signal is generated that is sensitive to displacement perpendicular to the longitudinal direction while the signal is insensitive to displacement along the longitudinal axis. The detector can thus detect a position displacement perpendicular to the longitudinal axis, while the detector is also displaced along the longitudinal axis.

In a particular embodiment, the first detector comprises a first and a second detector coil that are overlapping with half a coil-width. This makes it possible to generate two redundant position signals, of which one detects a maximum when the other detects a minimum in flux. This can be used to achieve fault-tolerance of the sensor and also to enhance the detection accuracy. . In addition, this makes detection insensitive to distance between the transmitter and detector coils.

In a particular embodiment, the second detector has a geometry that overlaps the second transmitter coil in a direction perpendicular to the longitudinal axis to enable generation of a position signal that is a function of the longitudinal displacement, but insensitive to displacement in the perpendicular direction.

In a further embodiment, the second detector comprises a third and a fourth detector coil which are overlapping with half a coil-width. Again, this makes it possible to obtain two position signals to yield enhanced accuracy of detection and ability to make a fault-tolerant construction. In addition, this makes detection insensitive to distance between the transmitter and detector coils.

In a further embodiment, the first transmitter coil is driven by a first electrical signal, the second by a sec^¬ ond electrical signal, such that the electromagnetic fields generated by the two transmitter coils can be distinguished from each other. This is used for both dis- crimination in the detectors and for elimination of electromagnetic disturbances.

In a further embodiment, the first and second transmitter-coils are mounted on a surface. This makes the coil construction very compact in a direction perpendicular to the said surface.

In a form where the said surface has the shape of a cylinder or a section thereof, the first and/or the second detector is mounted on a double-curved, eg a spherical surface. This construction allows to maintain an essentially constant distance between transmitter and detector coils when turning or displacing the objects relative to each other. Although not required for the proper func- tion, this will reduce effects of disturbing magnetic fields .

The invention is explained below with reference to the drawings, wherein

Fig. 1 shows a cylinder in a three-dimensional system of co-ordinates;

Fig. 2 shows a Meander-coil and a ring-coil; Fig. 3 is a sectional view through a number of electrical conductors that carry a current and thus generate an electromagnetic field;

Fig. 4 is a sectional view of two parallel conductors carrying a current and the thus generated electromagnetic field;

Fig. 5 is a sectional view of the Meander-coil and a de- tector with position signals in a detector made with two coils;

Fig. 6 shows two parallel conductors and a detector in the embodiment with two detector coils, and the position signals associated with a displacement;

Fig. 7 shows two detectors and their positions relative to the Meander and ring coils;

Fig. 8 shows a cylinder tilted around a z-axis and an x- axis and two detector-mounts;

Fig. 9 shown the locations of two detector-mounts relative to a cylinder ring in normal position;

Fig. 1 shows a cylinder ring in a three dimensional system of co-ordinates. The system of co-ordinates has an origo and three orthogonal axes, x, y and z. The cylinder ring can turn about the y and either or both of the x or z axes. Alternatively it can turn about y, rotate around x and/or z while also being displaced along y.

The cylinder ring is in the following used to illustrate the position of the first object relative to the second. The objects are not shown. The first object is connected with the cylinder ring, the second is fixed with the system of co-ordinates.

Fig. 2 shows a Meander coil and a ring coil. The Meander coils comprise a conductor in a Meander pattern 2 and a return conductor 3. The Meander shaped conductor is at one end terminated in A, in the other to the return conductor. The return conductor 3 is, in turn, connected to the Meander conductor and terminated in B. The return and the Meander conductors are mutually insulated, except for the one connection point. The Meander pattern has a number of Meander periods of length P, which is considered 360 Meander degrees or two half periods of 180 Meander degrees .

The ring coil comprises a ring-shaped conductor 4, which in each end is terminated in C and D. The ring coil is made to encompass the Meander coil and be electrically insulated from it.

The system of co-ordinates with the vertical y-axis and horizontal x- and z-axes show the orientation of the ring coil and the Meander coil.

The Meander and the ring coils have each an appropriate geometry as a coil system, to enable a well-defined spatial electromagnetic field. When the field is measured with an appropriate detector, it is possible to determine the position-relation by use of the detector. In the fol- lowing, a particular embodiment is considered where the cylinder ring 1 is mounted on a moving object. The Meander and ring coils are mounted on the surface of the cylinder ring 1, and the y-axes of the two bodies are parallel in a neutral position. A first electrical current i_M will flow through the Meander-coil when terminals A and B are connected to an appropriate source. The current creates an electromagnetic field around the conductors in the Meander coil. The field from the vertical conductor parts gives a total field component from the Meander that varies along the x- z plane. This component is used to detect rotation about the y-axis. The electromagnetic field from the horizontal conductors of the Meander gives rise to undesired compo- nent of the field, but with the return conductor placed in the mid-line of the Meander, the components from the return conductor and the horizontal parts will essentially give a zero component, since the orientation of these conductors are the same while they carry the same current but in opposite directions.

When the ring coil is connected via C and D to an electrical source, the current i_R will flow and an associated electromagnetic field will be generated. The field from horizontal parts of the coil creates a field component that varies along the y direction while, in turn, is invariant to rotation of the cylinder ring. This component is used for detection of rotation of the cylinder ring 1 around the x-axis and/or the z-axis, or for displacement along y. The electromagnetic field from the vertical conductor parts gives rise to undesired components of the field, but when the ring-coil spans exactly the entire circumference of the cylinder ring, the vertical parts of the coil can be placed isolated, but on top of each other such that the field disturbance from these parts will essentially cancel each other.

In a practical embodiment, the termination points are placed as close as possible, if coils are made as conduc- tive patterns on a flexible multi-layer printed circuit board, termination points can made to overlap without electrical short-circuiting. This effectively minimises termination effects.

Fig. 3 shows a cross section through a number of conduc- tors and the associated magnetic field. The field corre^¬ sponds to that established by the vertical conductors in the Meander 2 that is mounted on the cylinder 1. The cross section is seen in the x/z plane. The conductors 5 carry a current i_M shown as an x when the current is in the negative y direction. A dot indicates current out of the x/z plane in positive y direction. The direction of the magnetic field is illustrated by arrows.

The figure shows that along the circle 6, in the x/z plane and centered at the y axis, there is established a electromagnetic field that varies in a well-defined way as a function of the spatial position along the circle.

Fig. 4 shows a section of two parallel conductors carry- ing a current and generating an associated electromagnetic field. This illustrates the field generated by a current i_R in the conductors 7 and 8 in the ring coil 4. The generated field is shown as arrows, an x indicates direction perpendicular into the cut, a dot perpendicular out of the cut. There is hence established an electromagnetic field that varies in a well-defined way as a function of the spatial position along the y-axis while it is constant for a rotation around the y-axis.

Fig. 5 shows a section of the Meander coil and a detector, and the position signals. The conductor 9 has a cen- terline 10 and a geometric form with an extension hi along the y axis and Bl along the x/z axis. On the figure, Bl has an extension of 180 Meander degrees. The coils 11 and 12 are displaced relative to each other along the x/z axis by 90 Meander degrees. When a current flows in the conductor 9, a voltage ul is induced in the coil 11 and a voltage u2 is induced in the coil 12. The voltages ul and u2 will depend on the rela- tive displacement along the x/z axis, dxz, of the coils and the conductive elements. Voltages ul and u2 will, however, be invariant against a bounded, relative dis^¬ placement along the y axis since hi is less than HI. This means that by placing the coils 11 and 12 in close prox- imity to the surface of the cylinder 1, it is possible to detect a rotation of the cylinder 1 about the y axis despite this has also had a bounded rotation around another axis or a displacement along the y axis.

The voltages ul and u2 are functions of the relative dis^¬ placement dxz between the Meander coil and the coil 11 and the coil 12, respectively. With the coils 11 and 12 displaced by 90 Meander degrees, the phase angle between of ul and u2 have a similar difference.

If the current i_M is a direct current (DC) , the induced voltage in the coils 11 and 12 will be approximately sine and cosine of the displacement d, when Bl is chosen to be equivalent to 180 Median degrees. In a practical embodi- ment, alternating current will be chosen for i_M and the ul and u2 signals will be amplitude modulated according to the absolute value of the said functions of the displacement. Standard techniques are used to de-modulate the displacement information and eliminate disturbances from other sources.

Fig. 6 shows two parallel conductors, detector and position signals. The two conductors 13 and 14 are parallel and have a distance of h2. The coils 15 and 16 have a geometrical form with extension H2 along the y-axis and B2 along the x-z axis. When a current i_R flows through conductors 13 and 14, a voltage u3 is induced in the coil 15 and a voltage u4 in 16. The voltages u3 and u4 will depend on the relative displacement along the y-axis, dy, between the coils and the conductor. The voltages u3 and u4 will, however, be invariant against a relative displacement dxz along the x-z axis , between the coils and the conductor. With detector coils 15 and 16 placed in close proximity to the surface of the cylinder 1, it is thus possible to detect rotation of the cylinder 1 around the x and/or z-axes, independent of the rotation of the cylinder around the y- axis .

The voltages u3 and u4 are functions of the relative displacement dy between the ring coil and the coils 15 and 16, respectively. With 15 and 16 given a relative displacement by ^H2 = ^h2, voltages u3 and u4 have a difference in phase angle, such that one has a maximum when the other has its minimum.

If using a DC current for i_R , induced voltages in coils 15 and 16 will be approximately sine- and cosine functions of the displacement dy if H2 is chosen to equal the distance h2 between conductors 13 and 14.

Practical implementation will use a modulated (AC) signal in the ring coil and detection techniques to isolate the two signal envelopes which constitute the position sig- nals. The modulation type and frequency must be chosen from the maximum relative velocities of the detection coils relative to the transmitter coils using well known calculation techniques.

Fig. 7 shows two detectors relative to a ring coil and a Meander coil. The ring coil 18 and the Meander coil 17 are shown in dotted line, representing the ring coil 4 and the Meander coil 2 in Fig. 2. The detector 19 comprises two detector coils, like 11 and 12. The detector 20 likewise comprises two coils, like 15 and 16. In order to distinguish the signals from the ring and Meander coils, respectively, it is preferred to drive these coils from separate sources, using electrical excitation waveforms that enable separation of the two signals at the detector side. Several excitation waveforms can be used, including but not limited to pure sinusoidal at different frequencies, narrow band with adequate frequency separation, and spread spectrum with adequate coding.

Fig. 8 shows a cylinder, which is rotated relative to the two detectors, around a z-axis and an x-axis. This embodiment comprises two detector mounts 23 and 24 to enable measurement of rotations around the x-axis, the y- axis, and the z-axis. At least one of the detector mounts 23 or 24 shall comprise a detector, equivalent to the de- tector 19, to enable measurement of rotation around the y-axis. Rotation around both of x- and z-axes can be measured when both detector mounts comprise a detector equivalent to 20.

Fig. 9 shows the location of two detectors relative to a cylinder in normal position. Normal position is understood as the position of the cylinder where it is not rotated or displaced with respect to a system of coordinates defined in Fig. 1. Detectors 23 and 24 are mounted in close proximity to the inside of the cylinder 25. This is a preferred embodiment, although an outside mounting would work equally well.

The above description shows how the invention can be used for relative displacement, that is a measurement of position within one period of the Meander coil. In cases where absolute position of the Meander coil is needed, a second Meander coil system is established using a geome^¬ try with a different number of Meander periods in the first and second Meander coil. The combination of the two measurements will enable the resolution of absolute position.

With a first Meander of length L and N Meander periods, the relative position signal xi is related to absolute position as x= (x₁+n₁)-

where ni is an integer number in the closed interval from 0 to N-l, ni being the Meander period. With a second Meander coil, with M periods, the absolute position is x=(x2+"2)^

and the period count is n₂, an integer number in the closed interval from 0 to M-l.

Given Xi and x, L, N and M, these equations can be solved and the absolute position x be determined. It is a prerequisite that the design parameters N and M have no common divisors.

In the position measurement system as described in this invention, the implication is that absolute position measurement can be made in all degrees of freedom consid- ered, provided a second Meander coil is included.

In an implementation based on multi-. layer printed circuit board (PCB) technology, the Meander coils could conveniently be implemented in separate PCB layers. The above mentioned excitation principles should likewise be applied for a double Meander coil arrangement to enable convenient detection of absolute position.

Claims

C l a i m s

1. A system for measurement of the position of a first object relative to a second, comprising: a) a first inductive system comprising a transmitting coil and a first detector, with the attribute that the coil comprises a pattern of conductors perpendicular to a longitudinal axis, formed to make an electromag- netic field the flux of which, through a defined area, will vary by displacement of the conductor in the lon^┬¼ gitudinal direction, whereas the flux through the said area is essentially invariant to a displacement per^┬¼ pendicular to said direction.

b) a second inductive system comprising a second transmitting coil and a second detector, with the attribute that the coil comprises a conductor pattern essential in the longitudinal direction, which is formed to make an electromagnetic field the flux of which, through a defined area, will vary by displacement of the conductor perpendicular to the longitudinal direction whereas the flux is essentially invariant to a displacement in the longitudinal direction.

2. A system according to claim 1, c h a r a c t e r i z e d in that the first transmitter coil comprises a second conductor which is essentially parallel to the first conductor.

3. A system for position measurement according to claim 2, c h a r a c t e r i z e d in that the first and second conductors have a distance such that, when carrying an electrical current, the conductors generate an electromagnetic field that can be detected by the first detector.

4. A system for position measurement according to claims 1-3, c h a r a c t e r i z e d in that the first transmitter coil comprises a ring-shaped coil, which en- circles the second transmitting coil, as illustrated in Figures 1 and 2.

5. A system for position measurement according to claims 1-4, c h a r a c t e r i z e d in that the conduc- tors in the second transmitting coil are connected to form a pattern around a center-line, such that the flux of a generated electromagnetic field through a defined area will vary by displacement of the conductor in one direction but be essentially invariant to displacement perpendicular to said direction.

6. A system for position measurement according to claim 5, c h a r a c t e r i z e d in that the second transmitter coil comprises a Meander pattern coil around a centerline, as shown in Figure 2.

7. A system for position measurement according to claim 5-6, c h a r a c t e r i z e d in that the second transmission coil comprises a return conductor placed at the said center-line.

8. A system for position measurement according to claims 1-7, c h a r a c t e r i z e d in that the first detector has geometry and span to overlap the first transmitter coil in a direction perpendicular to longitudinal, whereby a longitudinal displacement of the transmitter coil will generate a position signal while the signal is invariant when the transmitter is displaced perpendicular to said direction.

9. A system for position measurement according to claims 1-8, c h a r a c t e r i z e d in that the first detector comprises a first and a second detector coil, the latter being displaced by half the width of the pat- tern of the first transmitter coil.

10. A system for position measurement according to claims 1-9, c h a r a c t e r i z e d in that the second detector has a geometry and span to overlap the second transmitter coil, whereby a position signal is induced when the transmitter is displaced in a direction perpendicular to longitudinal while the induced signal is invariant when the transmitter is moved in the longitudinal direction.

11. A system for position measurement according to claims 1-10, c h a r a c t e r i z e d in that the second detector comprises a third and fourth detector coils, the latter being displaced by half the height of the ring coil.

12. A system for position measurement according to claims 1-11, c h a r a c t e r i z e d in that the first transmitter coil is driven by a first electric sig- nal, and the second transmitter coil by a second electric signal such that the two fields generated by the said coils can be discriminated.

13. A system for position measurement according to claims 1-12, c h a r a c t e r i z e d in that the first and second transmitter coils are mounted at a surface.

14. A system for position measurement according to claim 13, c h a r a c t e r i z e d in that said surface is a cylinder and the first and/or second detector (s) is/are mounted on a double curved surface.

15. A system for position measurement according to claims 1-12, c h a r a c t e r i z e d in that two superimposed Meander coils are used instead of one, with the geometry of said coils made to give a non-integer ratio between the number of Meander periods of the two coils, and the two position signals from said coils are combined to give absolute position along the essential extent of the coil.