Dead Reckoning and Compass SensorsDead reckoning is the mathematical process of calculating the current position of an object based on knowledge of the location of its previous position and the distance and direction it travelled from this position to reach its current position. It is a navigational system which has been in use since medeval times and is very popular in robotics. Whilst deadreckoning is one of the simplest forms of navigation, implementation of the method suffers from an unbounded accumulation of errors. These errors can be significantly reduced if an absolute direction reference is available for the dead reckoning calculations.
Geomagnetic compass sensors provide an absolute direction reference. Provided there are no local sources of ferrous material or magnetic fields a freely suspended magnet will always point to the magnetic north pole. The Earth generates a magnetic field due to the motion of liquid layers which surrounds the Earth's iron core. The field surrounds the Earth and the north magnetic pole is located in the Parry Islands of northern Canada approximately 400 km from the axial north pole.
Large errors can be induced in geomagnetic compasses by the close proximity of a magnetic field or source of ferrous material. This is often stated as the main reason for not using these sensors. What follows is a description of a system which has been designed to identify and where possible remove this source of error. Geomagnetic Compass ErrorsThe magnetic field surrounding the Earth and the north magnetic pole is located approximately 400 km from the axial north pole but constantly moves. The resulting field is horizontal at the equator and vertical at the magnetic poles. In the absence of any local sources of ferrous material or magnetic fields a geomagnetic compass will always point to the magnetic north pole. Additional errors can also be induced in a geomagnetic compass by the vertical portion of the earth's magnetic field, these errors are greatest near the magnetic poles. The geographic location of the UK results in these errors being insignificant provided the compass sensors are kept horizontal.
The position of the magnetic pole with respect to the axial pole and the movement of the magnetic field produce a declination which varies with time and geographic location. The amount and variation with time of the declination for any given area is printed on Ordinance Survey maps and can easily be incorporated into dead reckoning software. The declination correction need only be applied if the heading information is to be related to true north rather than magnetic north.
Of greatest significance are the errors induced by the close proximity of ferrous objects or local magnetic fields such as those produced by electric cables or the magnets in motors. It is often stated that it is difficult to use geomagnetic compass sensors indoors due to the distortion from electric cables and steel structures and this is the main reason for not using a compass. Any deviations to the geomagnetic compass sensors due to the vehicle on which it is mounted can be reduced by the placement of compensating magnets or iron bars in the close proximity of the compass. This technique is described in detail in the old Admiralty Handbooks published by HMSO.
Ferrous objects, in particular large objects such as ships, gain a permanent magnetic field during construction due to the electric currents used during welding and the effects of hammering and heating the metal during riveting or forming the members. If the object is left in one position for a long period of time it will also gain an induced magnetic field due to the effects of the earth's magnetic field.
The deviation of any geomagnetic compass mounted on the vehicle will vary with the direction the vehicle is pointed and with latitude. The deviation may be reduced by placing small corrector magnets, soft iron spheres and soft iron bars known as flinder's bars around the perimeter of the compass. The process is one of trial and error which used to be carried out yearly on naval vessels. The Aldmiralty Handbook also required that the officer of the watch remove his keys, knife and any other magnetic objects from his person before using the ship's compass.
Whilst the full process of correcting a compass may be excessive for the application of compasses in robotics it does highlight the need to site the sensor carefully to avoid deviations and to choice the materials out of which the robot is constructed carefully. When mounting the compass sensors on the wheelchair chassis which was to be used as a mobile robotics platform it was found that the effect of the steel frame and motor magnets diminished significantly if the sensor was kept at least 150 mm from any such objects. As a result the sensors were mounted on a copper tubing frame at the front of the wheelchair.
Background reading failed to find any methods to directly reduce or eliminate the effect of errors sources other than those associated with the vehicle on which the sensors were mounted. Although several sources suggested that the fusion of the compass sensor data with data from other sensors may overcome the problem. External Error Source IdentificationThe possible sources of error due to external phenomena or objects identified during background research were magnetic fields associated with electric cables and the attraction of the geomagnetic compass sensor to ferrous metal objects such as heating radiators or structural steel members. Trials with a hand held compass designed for mountaineers suggested that such objects influenced the compass at a distance of approximately 300 mm.
A basic principle of physics was exploited to provide an innovative means of identifying and correcting for external error sources. Due to the distance travelled the light from the sun is assumed to be from a point source and all the light rays are assumed to be parallel to each other. Any source which is an infinite distance from a detector will produce radiation in parallel rays. If the magnetic north pole, towards which all geomagnetic compasses point, is assumed to be an infinite distance from the geomagnetic compass then this principle will also apply. Thus all geomagnetic compasses will point in the same direction regardless of the separation between the individual compasses. Whilst strictly this is not true, the declination variation (taken from an Ordinance Survey map) is 3.24" per meter in the area the project work was carried out. This figure is 2 orders of magnitude less than the typical manufacturer's stated accuracy of compass sensors and significantly less than the 10' annual change in declination.
To implement this principle a number of geomagnetic compass sensors need to be arranged such that it is not possible for an external ferrous source or magnetic field to effect all of the compass sensors in the same manner. If enough sensors are used it is then possible to use a voting system to choice the correct bearing to magnetic north if all of the sensors are not giving the same reading. With only two sensors it is possible for an external ferrous material source or magnetic field directly in line with the two sensors to produce a result similar to that of the magnetic north pole. By using a triangular formation it is not possible for an external magnetic field or source of ferrous material to produce the same deviation in all three compasses, but it is possible for two of the sensors to be indicating the same direction incorrectly and thus causing an error in the voting system. A minimum of five sensors arranged in a pentagon are required to ensure that a voting system will always indicate a true bearing to magnetic north, although this arrangement would still fail in the presence of a very large external magnetic field. The Resource constraints will usually limit the system to the use of three compass sensors arranged to occupy the points of a triangle. The length of the sides of the triangle are limited by the wheelchair frame footprint and are approximately 300 mm long. Figures below show the operation of the compasses under a variety of situations.
Figure 1
Figure 2
Figure 3
Figure 1 shows the direction indicated by the compass sensors when the robot is travelling towards magnetic north with all of the sensors pointing in the same direction. If the robot undertakes a 90o turn the three compass are all still pointing in the same direction as shown in figure 2. However if, as shown in figure 3, the compass sensors are influenced by an external ferrous material or magnetic field source at least one and possibly all three sensors will point to the error source. In order to produce an error the error source must be close to the robot in which case the distance from the source to the sensor can no longer be assumed to be infinite. As a result the sensors will not all give indicate the same direction and this can be detected by the compass sensor processing software, action can now be taken on discovering the discrepancy.
Local attractions such as buildings or metal objects can affect the accuracy of a prismatic compass survey by causing magnetic interference. This interference can lead to errors in compass readings and, consequently, inaccurate survey measurements. To minimize these effects, surveyors should identify and account for local attractions during their surveys, employing techniques such as taking multiple readings and using a declination correction.
The aims of a compass survey are to determine the magnetic bearing of a line and to set out accurate directions on the field. Its objectives include establishing a reference direction, determining the magnetic declination, and providing a basis for orienting maps and conducting navigation activities.
The local attraction are the error affecting device available at near by the compass while doing surveying. the local attraction is the effect of the magnetic materials on the compass and gets the incorrect bearing.
To use a prismatic compass, first ensure it is set up correctly by adjusting the azimuth ring to align with magnetic north. Then, hold the compass level and sight your target through the prism while reading the azimuth on the compass dial. Ensure you compensate for any declination angle if needed.
Measuring the bearing of survey lines at both ends ensures accuracy and reliability in determining the direction of the survey line. This helps in verifying and adjusting for any errors that may have occurred during the surveying process, ultimately leading to more precise results for mapping and construction purposes.
Local attractions such as buildings or metal objects can affect the accuracy of a prismatic compass survey by causing magnetic interference. This interference can lead to errors in compass readings and, consequently, inaccurate survey measurements. To minimize these effects, surveyors should identify and account for local attractions during their surveys, employing techniques such as taking multiple readings and using a declination correction.
1.useful in areas which are difficult to be accesed while chain survey is limited 2.less subjective to errors,in comparison to chain survey which subjected with errors such as change in chain length 3.useful in surveying a large area since it uses transverse(taking bearing from observer to an object ana vice versa is true) 4.it is accurate method for mapping,compare to chain survey
The aims of a compass survey are to determine the magnetic bearing of a line and to set out accurate directions on the field. Its objectives include establishing a reference direction, determining the magnetic declination, and providing a basis for orienting maps and conducting navigation activities.
sources of errors encountered in measurment
The local attraction are the error affecting device available at near by the compass while doing surveying. the local attraction is the effect of the magnetic materials on the compass and gets the incorrect bearing.
The errors of a magnetic compass include deviation, which is caused by local magnetic fields on the vessel, and variation, which is the difference between true north and magnetic north. Other errors can be caused by inclination, interference from metallic objects, and incorrect calibration. It is important to regularly check and adjust the compass to minimize these errors for accurate navigation.
To use a prismatic compass, first ensure it is set up correctly by adjusting the azimuth ring to align with magnetic north. Then, hold the compass level and sight your target through the prism while reading the azimuth on the compass dial. Ensure you compensate for any declination angle if needed.
The disadvantages of using a compass for surveying is that readings are often inaccurate (accuracy dependent on size and graduations of compass, it is easily disturbed by effects of local magnetic attraction
national sample survey
The US Navy while performing a careful survey of magnetic compass anomalies in the early 1950s to improve navigation. The data and the survey was originally classified.
Sampling and Non sampling errors
advantage it helps u know the direction of your measurement in relation to magnetic bearing disadvantage it is affected by the weather