System and method for calculating the orientation of a device

Abstract

A system and method for determining orientation of a vehicle is provided. The method includes the steps of providing a vehicle having a hinge joint such that sections of the chassis are capable of rotation with respect to each other. A first and second wheel is mounted to one and the other of the chassis sections, respectively. vehicle geometric data defining a distance between the hinge joint and the centers of the first and second wheels, respectively, and the diameter of the wheels is provided. surface geometric data defining the curvature of the surface can be provided. The angle of rotation about the hinge joint is measured. An orientation of the vehicle relative to the surface based on the vehicle geometric date, the surface geometric data, and the measured angle of rotation can be determined. A system and method for determining the orientation of an object is also provided.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This determine

Where J_F is the Jacobean matrix of the system F(X):

$J_F=\left[\begin{array}{cccccc}\frac{\partial F_1}{\partial x_1}& \frac{\partial F_1}{\partial y_1}& \frac{\partial F_1}{\partial x_2}& \frac{\partial F_1}{\partial y_2}& \frac{\partial F_1}{\partial x_h}& \frac{\partial F_1}{\partial y_h}\\ \frac{\partial F_2}{\partial x_1}& \frac{\partial F_2}{\partial y_1}& \frac{\partial F_2}{\partial x_2}& \frac{\partial F_2}{\partial y_2}& \frac{\partial F_2}{\partial x_h}& \frac{\partial F_2}{\partial y_h}\\ \frac{\partial F_3}{\partial x_1}& \frac{\partial F_3}{\partial y_1}& \frac{\partial F_3}{\partial x_2}& \frac{\partial F_3}{\partial y_2}& \frac{\partial F_3}{\partial x_h}& \frac{\partial F_3}{\partial y_h}\\ \frac{\partial F_4}{\partial x_1}& \frac{\partial F_4}{\partial y_1}& \frac{\partial F_4}{\partial x_2}& \frac{\partial F_4}{\partial y_2}& \frac{\partial F_4}{\partial x_h}& \frac{\partial F_4}{\partial y_h}\\ \frac{\partial F_5}{\partial x_1}& \frac{\partial F_5}{\partial y_1}& \frac{\partial F_5}{\partial x_2}& \frac{\partial F_5}{\partial y_2}& \frac{\partial F_5}{\partial x_h}& \frac{\partial F_5}{\partial y_h}\\ \frac{\partial F_6}{\partial x_1}& \frac{\partial F_6}{\partial y_1}& \frac{\partial F_6}{\partial x_2}& \frac{\partial F_6}{\partial y_2}& \frac{\partial F_6}{\partial x_h}& \frac{\partial F_6}{\partial y_h}\end{array}\right]$

Solving the above equation involves finding the inverse of the Jacobean, which can be time consuming and result in high demand of computing resources. Therefore the following system of linear equations can be used for faster results:
J_F⁻¹(X_n)(X_n+1−X_n)=−F_F(X_n)

For the first iteration, the vector Xn is the initial guess X₀. The initial guess used to ensure convergence towards a solution is the case where hinge point has the same y-coordinate as the driving wheel axis and angle hinge Φ is 180 degrees. In this case the initial guess becomes:

$X_0=\left[\begin{array}{c}{x}_1\\ y_1\\ x_2\\ y_2\\ x_h\\ y_h\end{array}\right]=\left[\begin{array}{c}L+D\\ b+R_1-R_2\\ L+D+W\\ b+R_1-R_2\\ L\\ b+R_1\end{array}\right]$

After solving the non-linear system, the hinge angle Φ is then calculated in terms of X as follows: ( )

$\varphi \left(X\right)=\frac{\pi }{2}+{\cos }^{-1}\frac{x_h}{R_1+b-y_h}+{\tan }^{-1}\frac{x_1-x_h}{y_h-y_1}+{\tan }^{-1}\frac{R_2}{D}$

A summary for the sequence of calculations in the forward model can be represented as follows:

##STR00001##

Inverse Model

In the inverse model, hinge angle Φ is given (measured using a potentiometer or other suitable sensor) and the vehicle orientation θ is to be calculated. In this case the major axis a is unknown. To calculate Φ, a system of seven non-linear equations can be constructed to solve the six unknowns in the model (x1, y1, x2, y2, xh, yh, a). These equations are the same six equations from the forward model plus the hinge angle equation Φ. The equations thus are:

$\frac{x_1^2}{a^2}+\frac{y_1^2}{b^2}=1\frac{x_2^2}{a^2}+\frac{y_2^2}{b^2}=1x_h^2+{\left(y_h-\left(b+R_1\right)\right)}^2=L^2{\left(x_2-x_1\right)}^2+{\left(y_2-y_1\right)}^2=W^2{x}_h=x_1-\frac{R\left(y_2-y_1\right)+D\left(x_2-x_1\right)}{W}{y}_h=y_1-\frac{R\left(x_2-x_1\right)-D\left(y_2-y_1\right)}{W}\varphi =\frac{\pi }{2}+{\cos }^{-1}\frac{x_h}{R_1+b-y_h}+{\tan }^{-1}\frac{x_1-x_h}{y_h-y_1}+{\tan }^{-1}\frac{R_2}{D}$ $F_l\left(X\right)=F_l\left(\left[\begin{array}{c}{x}_1\\ y_1\\ x_2\\ y_2\\ x_h\\ y_h\\ a\end{array}\right]\right)=\left[\begin{array}{c}\begin{array}{c}\frac{x_1^2}{a^2}+\frac{y_1^2}{b^2}-1\\ \frac{x_2^2}{a^2}+\frac{y_2^2}{b^2}-1\\ x_h^2+{\left(y_h-\left(b+R_1\right)\right)}^2-L^2\\ {\left(x_2-x_1\right)}^2+{\left(y_2-y_1\right)}^2-W^2\\ x_h-\left(x_1-\frac{R_2\left(y_2-y_1\right)+D\left(x_2-x_1\right)}{W}\right)\\ y_h-\left(y_1+\frac{R_2\left(x_2-x_1\right)-D\left(y_2-y_1\right)}{W}\right)\end{array}\\ \varphi -\left(\frac{\pi }{2}+{\cos }^{-1}\frac{x_h}{R_1+b-y_h}+{\tan }^{-1}\frac{x_1-x_h}{y_h-y_1}+{\tan }^{-1}\frac{R_2}{D}\right)\end{array}\right]=0$

This system can also be solved using Newton Raphson method and the initial guess is:

$X_0=\left[\begin{array}{c}{x}_1\\ y_1\\ x_2\\ y_2\\ x_h\\ y_h\\ a\end{array}\right]=\left[\begin{array}{c}L+D\\ b+R_1-R_2\\ L+D+W\\ b+R_1-R_2\\ L\\ b+R_1\\ b\end{array}\right]$

After solving the non-linear system, the vehicle orientation θ can be calculated in terms of X as follows:

$\theta ={\cos }^{-1}\left(\frac{b}{a}\right)$

A summary for the sequence of calculations in the forward model can be represented as follows:

##STR00002##

In order to To determine when every loop of the helical path ends and the next one starts, a combination of sensor readings can be used including wheel encoders and IMU data (e.g., data obtained from accelerometers, gyroscopes, and magnetometers). The localization scheme used relies on the pipe orientation with respect to the ground because it affects which sensors provide useful readings. If the pipe 80 is horizontal, as shown in FIG. 8A, then the accelerometers will provide useful data in determining the angular position of the vehicle 10 on the pipe. The gravity direction and thus the accelerometers readings will continuously change while vehicle is travelling around the pipe. Having such a variation aids the controller to know when the robot reaches, for example, the 12 o'clock position and register the end of a helix loop and the start of the next one. If the pipe 80 is vertical, as shown in FIG. 8B, then the accelerometers will not be useful because gravity direction with respect to the vehicle 10 will not change. Therefore, the accelerometers data will not change while the robot moves around the pipe regardless of which orientation the accelerometers are mounted on the robot. This is because the same face of the robot always faces the ground when the robot is running around a vertical horizontal.

Therefore, in the case of a horizontal pipe, the master microcontroller can depend on the accelerometer values to determine the end and start of the helix loops. On an inclined pipe, the accelerometer values can be used as long as the inclination is not close to being vertical; because if the pipe is vertical or close to vertical then the accelerometer values will stay constant or their changes will not be sufficient for meaningful determinations. In this case, other means can be used including magnetometer readings, dead reckoning using wheel encoder values, and/or monitoring external visual references. A Kalman filter or other suitable sensor fusion algorithm can be used to fuse these data for the case of a vertical pipe.

Accordingly, the above equations provide an exemplary method for determine orientation/hinge angle of a vehicle on a curved surface. The method can be adapted to different structures of vehicles/mechanisms that are capable of self-adjusting to a known curvature in a measurable way (e.g., having a hinge that is measured).

The method for determining the orientation of the vehicle can be performed by a computer having a processor with memory for executing code. As shown in FIG. 4 the functional elements of an information processor 102 are illustrated, and preferably include one or more central processing units (CPU) 202 used to execute software code in order to control the operation of information processor 102, read only memory (ROM) 204, random access memory (RAM) 206, one or more network interfaces 208 to transmit and receive data to and from other computing devices across a communication network, storage devices 210 such as a hard disk drive, flash memory, CD-ROM or DVD drive for storing program code, databases and application code, one or more input devices 212 such as a keyboard, mouse, track ball and the like, and a display 214. The various components of information processor 102 need not be physically contained within the same chassis or even located in a single location. For example, as explained above with respect to databases which can reside on storage device 210, storage device 210 may be located at a site which is remote from the remaining elements of information processors 102, and may even be connected to CPU 202 across communication network 106 via network interface 208. For example, data processing can be performed using processors located on board the robot and transmitted to a remote computer terminal.

The functional elements shown in FIG. 4 (designated by reference numbers 202-214) are preferably the same categories of functional elements preferably present in user computing device 104. However, not all elements need be present, for example, storage devices in the case of PDAs, and the capacities of the various elements are arranged to accommodate expected user demand. For example, CPU 202 in user computing device 104 may be of a smaller capacity than CPU 202 as present in information processor 102. Similarly, it is likely that information processor 102 will include storage devices 210 of a much higher capacity than storage devices 210 present in workstation 104. Of course, one of ordinary skill in the art will understand that the capacities of the functional elements can be adjusted as needed.

For example, sensors measuring the angle of the hinge can provide electrical input signals to the processor. Such signals can undergo analog or digital signal processing before being inputted to the processor 202, such as by a pre-processing module implemented as computer code. Such a module can receive output from an analog-to-digital converter, which in turn receives signals from a sensor, e.g., a strain gauge. The calculations used to determine the orientation of the vehicle can be performed by processors located on board the robotic vehicle. Alternatively, or in addition, sensed data can be transmitted (e.g., through wireless communications) to a remote processor (e.g., a field laptop computer, smartphone, tablet, etc.) to perform the processing to determine the orientation and location of the vehicle.

Determining the orientation of the vehicle is particularly useful in robotic inspection applications. For example, the orientation information can be used to calculate a trajectory of the vehicle as it travels along the surface. This can be used to determine the absolution absolute location of the vehicle. For example, to orientation information can be combined with a measured distance traveled by the vehicle (e.g., by counting rotations of the drive wheel) to determine where along the surface the vehicle is located. In addition, the orientation information can be particularly useful when the vehicle is being used to inspect the surface itself and the vehicle needs to pass over the surface to inspect it. The orientation information can be used to determine whether a desired sweep pattern has been achieved by the robot. For example, the distance and orientation information can be collected as data points and combined to build a map of travel of the vehicle. The map can be a three-dimensional map (e.g., using a cylindrical coordinate system) or can be displayed as a two-dimensional map by converting the cylindrical surface into a flat planar surface. Further, since the trajectory of the robot can be mapped, that location information can be combined with inspection data collected, to create detailed maps in which a map of the structure is overlaid with the inspection data. Thus, data points can be generated that include location information and inspection information (e.g., condition of the surface at that location). The data points can be used to create a detailed map of a pipeline in which areas of corrosion are highlighted on the map. Without the system and method of the present invention, the localization data required to produce such maps would be based in relative referencing with drifting errors which would accumulate over time.

Incorporating the system and method of the present invention in robotic vehicle 10, for example, can provide significant benefits. The system and method can be used for measuring the angle of the hinge as the robot maneuvers on a pipe and permits the absolute orientation of the robot with respect to the centerline (or the flow) of the pipe to be accurately calculated. This method can be extrapolated to other mechanisms, vehicles and curved surfaces. This method is unique in that it out performs IMUs in the particular case where the direction of the gravitational force around the vehicle is aligned with the axis of rotation around which the vehicle is pivoting. This alignment between the gravitational force and the axis of rotation can occur, for example, in a condition in which the robot located at the 12 o'clock position on a horizontal pipe and the robot is pivoting/steering in circles in place.

As discussed above, this method has particular application in the control of robotic inspection vehicles. For example, if a robot were to advance along a cylindrical pipe, a particular measurement of a hinge angle can correspond to four possible orientations of the robot relative to the pipe centerline or flow. However, if the original orientation of the robot was known, the system can be configured to store that date data and further store data indicating all further orientation changes. As such, the system can keep track of previous measurements of the hinge angle and easily deduce which one of those four possible current orientations is the true one based on the starting orientation and previous orientation changes.

In addition, if the geometry of the robot is known and the orientation of the robot relative to a pipe is known, the measured angle about the hinge can be used to determine the diameter of the pipe. Further, a robot can be placed on a pipe of unknown diameter and then commanded to pivot 181 degrees around its own driving wheel and map the data acquired from the sensor measuring the angle about the hinge. The system can record the maximum angle as the robot rotates, which occurs when the robot is oriented normal to the flow of the pipe (i.e. as if the robot was about to start driving in circles around the pipe circumference). The maximum measured angle, combined with the geometer data of the robot, can be used for determining the diameter of the pipe upon which the robot is traveling. The method can be adapted to different structures of vehicles/mechanisms that are capable of self-adjusting to a known curvature in a measurable way (e.g., having a hinge that is measured). A vehicle could be used to inspect ship hulls, including aerospace and marine hulls. In addition, if such an inspection tool were used on a structure that had a unique curvature over its circumference, the tool's position could be determined by measuring a hinge angle. For example, if the curvature of a submarine were unique across its hull and was known (e.g., could be represented graphically as a parabola), a measured hinge angle could be used to calculate the vehicle orientation and position.

As discussed above, the described technique could be used to vehicles and other mechanisms more generally. For example, a hinged pincer could employee the described methods for determining the orientation of an object being held by the pincers.

The method described above can provide useful data on applications where IMUs would not provide similar information. For example, smart grippers in automated assembly lines (e.g., pincers consisting of first and second prongs connected by a hinge) can be constructed according to the principles of this invention to grab scrambled objects (e.g., a pile of objects in a pile having random orientations), self-adjust to their shape and use the measured changes in the body of the gripper to determine which way the object was grabbed. For example, FIGS. 9A-9H illustrate several embodiments of grippers that can be used to determine the orientation of an object.

Referring to FIG. 9A, a gripper 910 has two legs 912 and a third leg 914 connected to the gripper via a pivot 916. The ends of the legs 912 and leg 914 include gripping elements 918 that contact the surface of an object. The object can have either a concave, convex, or flat surface, or combinations thereof. The gripping elements 918 maintain the gripper in contact with the object. The gripper elements can maintain contact with the object via magnetic force, suction force, non-permanent adhesion force, or other suitable means for maintaining the ends of the gripper in contact with the object. The geometric details of the gripper 910 and its legs and gripping elements are known. The geometric details of the object are also known. Accordingly, by measuring the amount of rotation at pivot 916 in accordance with the methods described above the orientation of an object being held by the gripper can be determined. As shown in FIG. 9E, gripper 910 is shown in contact with object X at various orientations along its surface. In accordance with the methods discussed above, the orientation of the object X relative to the gripper can be determined.

Referring to FIG. 9B, a gripper 920 has two legs 922 each connected to the gripper via a pivots 926. The ends of the legs 922 include gripping elements 928 that contact the surface of an object. The object can have either a concave, convex, or flat surface, or combinations thereof. The gripping elements 928 maintain the gripper in contact with the object. The gripper elements can maintain contact with the object via magnetic force, suction force, non-permanent adhesion force, or other suitable means for maintaining the ends of the gripper in contact with the object. The geometric details of the gripper 920 and its legs and gripping elements are known. The geometric details of the object are also known. Accordingly, by measuring the amount of rotation at pivots 926 in accordance with the methods described above the orientation of an object being held by the gripper can be determined. As shown in FIG. 9F, gripper 920 is shown in contact with object X at various orientations along its surface. In accordance with the methods discussed above, the orientation of the object X relative to the gripper can be determined.

Referring to FIG. 9C, a gripper 930 has two legs 932 each connected to the gripper and an element 934 that is capable of linear displacement. The ends of the legs 932 and element 934 include gripping elements 938 that contact the surface of an object. The object can have either a concave, convex, or flat surface, or combinations thereof. The gripping elements 938 maintain the gripper in contact with the object. The gripper elements can maintain contact with the object via magnetic force, suction force, non-permanent adhesion force, or other suitable means for maintaining the ends of the gripper in contact with the object. The geometric details of the gripper 930 its legs, translation element, and gripping elements are known. The geometric details of the object are also known. Accordingly, by measuring the amount of displacement of element 934 in accordance with the methods described above, wherein the rotation of a pivot is substituted for linear displacement, the orientation of an object being held by the gripper can be determined. The displacement of element 934 can be measured by a sensor, such as a spring strain gauge 936, for example. As shown in FIG. 9G, gripper 930 is shown in contact with object X at various orientations along its surface. In accordance with the methods discussed above, the orientation of the object X relative to the gripper can be determined.

Referring to FIG. 9D, a gripper 940 has two legs 942 each connected to the gripper. The ends of the legs 942 include gripping elements 948 that contact the surface of an object. The object can have either a concave, convex, or flat surface, or combinations thereof. The gripping elements 948 maintain the gripper in contact with the object. The gripper elements can maintain contact with the object via magnetic force, suction force, non-permanent adhesion force, or other suitable means for maintaining the ends of the gripper in contact with the object. The gripper includes a non-contact sensor 946 (e.g., ultrasound, light, laser, etc.) that can measure the distance between the sensor 946 and the surface of the object in line with the sensor. The geometric details of the gripper 930 its legs and the location of the sensor. The geometric details of the object are also known. Accordingly, by measuring the distance between the sensor 946 and the surface of the object in accordance with the methods described above, wherein the rotation of a pivot is substituted for distance measured by the sensor, the orientation of an object being held by the gripper can be determined. As shown in FIG. 9H, gripper 940 is shown in contact with object X at various orientations along its surface. In accordance with the methods discussed above, the orientation of the object X relative to the gripper can be determined.

FIGS. 9A-H illustrate embodiments in which a device (e.g., grippers 910, 920, 930, or 940) can be used to determine the orientation of an object that is in contact with the device. The method fixed geometric data concerning the device, fixed geometric data concerning the object, and at least one variable (e.g., single pivot rotation, double pivot rotation, linear translation, or distance) to determine the orientation of the object relative to the device.

It should be understood that various combination, alternatives and modifications of the present invention could be devised by those skilled in the art. The present invention is intended to embrace all such alternatives, modifications and variances that fall within the scope of the appended claims.

While the invention has been particularly shown and described with reference to a preferred embodiment thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

Claims

1. A method for determining orientation of a vehicle relative to a surface having defined geometric data that represents a curvature of the surface, the method comprising the steps of:

measuring an angle of rotation about a hinge joint that is part of the vehicle, wherein the vehicle is configured to travel along the surface;

determining the orientation of the vehicle relative to the surface based on vehicle geometric data, the surface geometric data and the measured angle of rotation using a processor located on board the vehicle; and

controlling operation of the vehicle in view of the determined orientation using the processor,

wherein the surface comprises a cylindrical pipe and the step of determining the orientation of the vehicle corresponds to determining one orientation out of four possible orientations relative to a centerline of the cylindrical pipe and the method further includes the step of storing an original orientation of the vehicle and all further orientation changes of the vehicle and determining which one of the four possible orientations is a true orientation of the vehicle based on the original orientation and all the further orientation changes of the vehicle.

2. The method of claim 1, wherein the hinge joint is disposed between a first chassis section and a second chassis section such that the first and second chassis sections are capable of rotation with respect to each other in at least one direction.

3. The method of claim 2, wherein the first chassis section includes a first wheel mounted thereto and the second chassis section includes a second wheel mounted thereto.

4. The method of claim 3, wherein the vehicle geometric data defines at least a distance between the hinge joint and a center of the first and second wheels, respectively, and a diameter of each of the wheels.

5. The method of claim 3, wherein the first wheel is a magnetic drive wheel.

6. The method of claim 3, wherein the second wheel is a magnetic drive wheel that is configured to move in at least two directions.

7. The method of claim 1, wherein the step of measuring the angle of rotation about the hinge joint comprises providing a sensor in the vehicle, the sensor being configured to measure the angle of rotation about the hinge joint.

8. The method of claim 1, wherein the vehicle is configured to transverse a curved surface in a helical path.

9. The method of claim 8, further including the step of determining when every loop of the helical path ends and a next loop starts by analyzing a plurality of sensor readings including wheel encoders and inertial measurement unit (IMU) data.

10. The method of claim 9, wherein the IMU data includes data obtained from accelerometers, gyroscopes and magnetometers.

11. The method of claim 1, wherein the surface comprises a cylindrical pipe and the step of determining the orientation of the vehicle corresponds to determining four possible orientations relative to a centerline of the cylindrical pipe and the method further includes the step of recording previous measurements of the hinge angle and determining which of the four possible orientations is a true orientation of the vehicle based on a start orientation and previous orientation changes of the vehicle.

12. The method of claim 8, wherein the vehicle includes a on board processor that operates is further configured to operate so as to control a helical pitch and ensure that loops of the helix are spaced a constant distance from one another.

13. The method of claim 3, wherein the first wheel is a magnetic drive wheel that rotates about a first axis and the second wheel is a magnetic omni-wheel that rotates about a second axis that is perpendicular to the first axis, the omni-wheel including a plurality of rollers that are located about a periphery of the omni-wheel, each roller rotating in a same direction as rotation of the magnetic drive wheel.

14. The method of claim 1, wherein the hinge includes first and second rotation stops that are configured to prevent unintentional rotation about the hinge.

15. The method of claim 14, wherein the hinge joint is disposed between a first chassis section and a second chassis section such that the first and second chassis sections are capable of rotation with respect to each other in at least one direction, wherein the first and second rotation stops are mating surfaces on each of the first and second chassis section sections, respectively.

16. A method for determining a diameter of a curved structure over which a vehicle travels comprising the steps of:

determining geometric data of the vehicle;

determining an orientation of the vehicle relative to the curved structure; and

measuring an angle of rotation about a hinge joint that is part of the vehicle to determine the diameter of the curved structure.

17. The method of claim 16, further including the step of determining a maximum angle of rotation that has been measured and the step of determining the diameter of the curved structure is based on the recorded maximum angle of rotation and the geometric data of the vehicle.

18. The method of claim 16, wherein the curved structure comprises a pipe.

19. A system for determining orientation of a vehicle relative to a surface having defined geometric data that represents a curvature of the surface comprising:

a drivable vehicle including:

a hinge joint connecting a first part of the vehicle to a second part of the vehicle such that the first and second parts are capable of movement with respect to one another as the vehicle is driven across the surface, the movement of the first and second parts being translated into rotation of the hinge joint;

a sensor configured to measure an angle of rotation about the hinge joint; and

a processor configured to determine an orientation of the vehicle relative to the surface based on defined vehicle geometric data, the defined geometric data of the surface, and a measured angle of rotation about the hinge joint and wherein the processor is configured to control operation of the vehicle in view of a determined orientation,

wherein the surface comprises a cylindrical pipe, and

wherein to determine the orientation of the vehicle, the processor is further configured to determine one orientation out of four possible orientations relative to a centerline of the cylindrical pipe by storing an original orientation of the vehicle and all further orientation changes of the vehicle and determining which one of the four possible orientations is a true orientation of the vehicle based on the original orientation and all the further orientation changes of the vehicle.

20. The system of claim 19, wherein the hinge is selected from the group consisting of: a knuckle/pin hinge, a ball and detent hinge, and a length of flexible material.

21. The system of claim 19, wherein the first part comprises a first chassis section and the second part comprises a second chassis section.

22. The system of claim 21, wherein the first chassis section includes a first wheel mounted thereto and the second chassis section includes a second wheel mounted thereto.

23. The system of claim 19, wherein the defined vehicle geometric data includes at least a distance between the hinge joint and a center of the first and second wheels, respectively, and a diameter of the wheels.

24. The system of claim 19, wherein the first wheel comprises a magnetic drive wheel and a second wheel comprises a magnetic drive wheel.

25. The system of claim 19, wherein the vehicle is capable of traversing a curved surface in a helical pattern.