There exists a need for a robot having the ability to navigate obstacles, climb, and turn over and within areas, such as the hull of a vessel, the interior of a tank, and difficult to reach passageways that feature geometric discontinuities caused by, for example, plumbing, protrusions, and indentations.
Reference in the specification to “one embodiment” or to “an embodiment” means that a particular element, feature, structure, or characteristic described in connection with the embodiments is included in at least one embodiment. The appearances of the phrases “in one embodiment”, “in some embodiments”, and “in other embodiments” in various places in the specification are not necessarily all referring to the same embodiment or the same set of embodiments.
Some embodiments may be described using the expression “coupled” and “connected” along with their derivatives. For example, some embodiments may be described using the term “coupled” to indicate that two or more elements are in direct physical or electrical contact. The term “coupled,” however, may also mean that two or more elements are not in direct contact with each other, but yet still co-operate or interact with each other. The embodiments are not limited in this context.
As used herein, the terms “comprises,” “comprising,” “includes,” “including,” “has,” “having” or any other variation thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, article, or apparatus that comprises a list of elements is not necessarily limited to only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Further, unless expressly stated to the contrary, “or” refers to an inclusive or and not to an exclusive or.
Additionally, use of the “a” or “an” are employed to describe elements and components of the embodiments herein. This is done merely for convenience and to give a general sense of the invention. This detailed description should be read to include one or at least one and the singular also includes the plural unless it is obviously meant otherwise.
The embodiments disclosed herein describe a segmented system with magnetic wheels that can, for example, navigate the hull, tanks, and passageways of a ship. The system provides effective climbing and turning ability over and within a ferrous hull that typically features geometric discontinuities in the form of plumbing, protrusions, and indentations, such as weld seams where hull plating meets.
Linkage arm 20 helps transfer push (compressive) and pull (tension) forces between the drive modules 30 and 40 so they work together in concert to overcome obstacles greater than the capability of any one drive module. In some embodiments, linkage arm 20 is arched, while in other embodiments linkage arm 20 is straight (i.e. aligned in height with drive modules 30 and 40). In some embodiments, linkage arm 20 comprises a rigid material, while in other embodiments linkage arm 20 comprises a flexible material, such as rubber, to allow relative motion between drive modules 30 and 40 so that system 10 can turn, negotiate obstacles, and traverse around corners.
A flexible linkage arm 20 allows for all three rotational degrees of freedom (roll, pitch, and yaw), providing excellent mobility at the expense of vehicle control. Accordingly, in such embodiments, a sensor system that provides for accurate relative positioning between drive modules 30 and 40 may be used. One example sensor is a set of stereo cameras on each drive module 30 and 40 that watch the drive module in front of it. The stereo vision output from the sensor can be used to provide accurate pose between drive modules allowing them to coordinate motion for maximum mobility.
In some embodiments, linkage arm 20 is semi-rigid, meaning that linkage arm 20 is comprised of a rigid material but is configured to have a specific degree of freedom (DOF) such as shown in
As an example, hinged connection 26 may be made as shown in
In some embodiments, and as discussed in more detail with respect to
In some embodiments, and as discussed more in detail with regard to
In some embodiments, first drive module 30 and second drive module 40 are independently controlled, such as via a remote controller (not shown). In some embodiments, first drive module 30 and second drive module 40 are controlled as a whole system using a control algorithm. In some embodiments, first drive module 30 and second drive module 40 contain accelerometers therein to increase control accuracy. Magnetic wheels 32 and 34 are connected to first drive module 30 and magnetic wheels 42 and 44 are connected to second drive module 40. As an example, magnetic wheels 32, 34, 42, and 44 may be configured similarly to magnetic wheels 300 shown and described with reference to
Linkage arm 110 comprises a first linkage arm portion 112 having at least one degree of freedom with respect to a connected second linkage arm portion 114. As an example, and as shown in
First linkage arm portion 112 and second linkage arm portion 114 are connected via a hinged connection that allows for lateral movement with respect to the direction of orientation of linkage arm 110 with respect to first drive module 150 and second drive module 160.
For the connection of the other end of first linkage arm portion 112 to first drive module 150, roll shaft 126 is passed through an opening in the end of linkage arm portion 112 and secured by retaining ring 128. A washer 130 is placed over the distal end of roll shaft 126, with the distal end of roll shaft 126 passed through sealed bearings 132, which are press-fit into encoder chassis mount 124, followed by washer 134, shaft retaining clip 136, encoder mounting plate 138, and roll encoder 140, with washer 134, clip 136, plate 138, and encoder 140 all fitting within mount 124 to ensure surface-to-surface contact between mount 124 and first drive module 150 and to allow smooth roll DOF movement.
Referring to
The DOFs are desirable for the platform to navigate real-world structures. The multiple DOFs allow the maximum wheel contact to the surface regardless of obstacles and surface curvature. In some embodiments, each DOF is outfitted with an encoder to provide pitch, roll, and/or yaw information. The coordinating controller uses the data as inputs to the PID controller (see
As shown by the arrow in
Referring to
A motor assembly 230 is configured to be housed within each end of drive module 200 to connect magnetic wheels 260 and 262 to drive module 200. Motor assembly 230 provides the torque to rotate the magnetic wheels and move the system. The primary design considerations for the motor assembly include torque output, speed output, shock absorption, weight, robustness, and modularity. Required motor torque may be calculated by multiplying the weight of a drive module segment by the magnetic wheel radius and safety factor of two. As an example, the required torque for a single motor output is 20 inch-pounds. A required speed output of 57 rpm is determined by multiplying the desired climbing speed of the system by the wheel circumference. These values may be used to select an optimal motor and gearbox combination for particular system requirements.
Most unmanned ground vehicles are submitted to large shock loads as the system traverses over rough ground. Systems that climb or are thrown tend to experience even larger shock loads in the inevitable case where they fall and hit the ground. Gearbox output shafts are often hardened and will easily break if a large radial load is experienced. As such, the design of drive module 200 mitigates the effects of large shock loads where the system falls onto the wheels. Wheel shaft 242 is supported by two high-load bearings 250 directly coupled to the housing 232.
If the system falls and impacts a wheel, the radial loads are distributed through wheel shaft 242 to housing 232 and back to housing 210 instead of to wheel shaft 242. Motor-to-wheel coupler 240 allows relative motion between wheel shaft 242 and the gearbox shaft during impacts that cause wheel shaft 242 to deflect, while allowing torque to be transmitted from shaft to shaft for vehicle motion.
Referring now to
Magnetic wheel assembly 300 provides the attractive force between the system and the surface traversed, allowing the system to traverse vertical and inverted ferrous surfaces. The outer surface of wheel assembly 300 has a high coefficient of static friction. Such a feature makes the attached system useful for exploration of ships, shipping containers, and other ferrous environments.
Wheel assembly 300 includes a wheel housing 310. Wheel housing 310 allows the entire wheel assembly 300 to flex during impacts with the ground. As an example, wheel housing may be made of rubber, such as neoprene, or other durable material that provides desired traction on a surface, or comprise a combination of materials having different characteristics. Wheel housing 310 includes a plurality of slots 312 disposed around the perimeter where magnets, such as N52 neodymium magnets, are inserted to provide for magnetic capability. In some embodiments, the magnets are inserted into the slots such that all like-poles are facing the same direction. This configuration offers the optimal magnetic flux at the contact points of wheel assembly 300 to the surface. The slots are oriented such that the array of magnets is oriented parallel to the central axis of wheel assembly 300.
A wheel core 316 is press-fit into wheel housing 310 for attachment to wheel shaft 242 shown in
Optimization of wheel assembly 300 using flux-plate wheel design is discussed in depth in a paper entitled “Design and Optimization of a Magnetic Wheel for Hull Climbing” by Kerber et al., the entire content of which is incorporated by reference herein. As an example, one embodiment of a wheel assembly 300 may be 1.25 inches wide, 4 inches in diameter, and have a measured attraction force of 21 lbf.
The idea behind the ball-joint linkage is that for compressive force to be translated from one robot module to the next the ball-joint must be locked in the pitch DOF or the linkage will simply start to fold on itself. The ball-joint design allows the magnetic wheels to maintain surface contact with pitch, roll, and yaw DOFs. The design also allows tension (pull) forces through the ball joint. Inside the ball joint, spring 418 pushes ball 416 into the normal operating position, but when a compressive force between drive modules 430 and 440 starts to compress spring 418, ball 416 slides further into its receptacle and the pitch-and-yaw DOFs are constrained. As the compressive force diminishes, spring 418 returns to its original position and the joint regains its full range of motion.
As shown, linkage arms 510 and 540 are arched. However, linkage arms 510 and 540 may comprise other orientations as would be recognized by one having ordinary skill in the art. Further, linkage arms 510 and 540, while shown as rigid, may be configured to have a hinged connection or other connection allowing for a degree of freedom between its respectively attached drive modules, such as shown in
In some embodiments, first linkage arm 510 has at least one DOF with respect to first drive module 520. In some embodiments, second linkage arm 540 has at least one degree of freedom with respect to third drive module 550. As an example, the degree of freedom may be provided by securing the respective linkage arm to the respective drive module using a roll shaft as shown in
System 610 contains two linkage arms, one of which, linkage arm 612 is flexible. The flexible linkage arm improves the ability of system 600 to negotiate corners of structures, such as structure 620.
The climbing operation of a two-segment system, such as system 500, is shown in
Software for the embodiments of the systems disclosed herein can be divided into driver module, operator control, and coordinated control. Driver module software runs on integrated circuits within the driver module and is responsible for low-level functionality of the system such as motor control and battery-level monitoring. The coordinated control software (CCS) receives feedback from the motor encoders and relative positioning sensors and provides the commands to the motor controllers for optimal motion along the system's path.
The CCS coordinates the motion of the drive modules to allow the system to climb, turn, and scale obstacles more effectively. Using the CCS, each wheel on the system is actuated individually, providing its own output speed. Although control of a four-wheel drive system using a standard radio controller (RC) is possible, such as system benefits from use of a coordinated controller. In some embodiments, a leader-follower control scheme may be used, where the operator directs the front drive module to go forward, left, or right. The rest of the drive modules will then follow the drive module in front of them using a simple proportional, integral, and derivative (PID) controller, such as that shown in the block diagram 800 in
As shown in diagram 800 of
Commanded user input 820 may include speed reference data 822 and rotation reference data 824. The difference between the motor speed encoder data 812 and the speed reference data 822 is provided to controller 830 and is multiplied by value KP 832, the integral of the difference value is taken and is multiplied by value KI 834, and the derivative of the difference value is taken and is multiplied by value KD, 836. Values 832, 834, and 836 are then summed.
Similarly, the difference between linkage rotation encoder data 814 and rotation reference data 824 is provided to controller 830 and is multiplied by value Kp 838. The integral of the difference value is taken and is multiplied by value KI 840, and the derivative of the difference value is taken and is multiplied by value KD, 842. Values 838, 840, and 842 are then summed. Both of the resulting summed values are then summed. The resultant value is then provided by controller 830 as motor input 850, which is used by the motor assembly, such as motor assembly 230 shown in
If the leader-follower controller is inadequate because of the pitch and roll DOFs, or if the linkage design is overly complex to support the leader-follower approach (e.g. design contains a plurality of segments), a more complex control architecture may be used. In such embodiments, the controller as shown in
In terms of kinematics, consider the robot motion on a planar magnetic surface, shown in
q=[x1y1θ1θ2]T (Eq. 1)
where x1 and y1 are the coordinates of the front module center of mass, assumed located in the front drive module's geometric center, and θ1 and θ2 are the orientations of the front and rear modules, respectively. The position of the rear module, (x2, y2), is related to the position of the front module by
x2=x1+d1 cos θ1+d2 cos θ2
y2=y1+d1 sin θ1+d2 sin θ2 (Eq. 2)
where d1 and d2 are the distances from the front and rear module centers of mass to the yaw joint, respectively. Differentiating Eq. 2 yields the linear velocities, ({dot over (x)}2,{dot over (y)}2), of the second module:
{dot over (x)}2={dot over (x)}1−{dot over (θ)}1d1 sin θ1−{dot over (θ)}2d2 sin θ2
{dot over (y)}2={dot over (y)}1+{dot over (θ)}1d1 cos θ1+{dot over (θ)}2d2 cos θ2 (Eq. 3)
where {dot over (x)}1 and {dot over (y)}1 are the linear velocities of the front module, and {dot over (θ)}1 and {dot over (θ)}2 are the rotational velocities of the first and second modules, respectively. Differentiating Eq. 2 again yields the linear accelerations, ({umlaut over (x)}2, ÿ2), of the second module:
{umlaut over (x)}2={umlaut over (x)}1−{umlaut over (θ)}1d1 sin θ1−{dot over (θ)}12−d1 cos θ1−{umlaut over (θ)}2d2 sin θ2−{dot over (θ)}22d2 cos θ2
ÿ2={umlaut over (θ)}1d1 cos {umlaut over (θ)}1−{dot over (θ)}12−d1 sin θ1+{umlaut over (θ)}2d2 cos θ2−{dot over (θ)}22d2 sin θ2 (Eq. 4)
where {umlaut over (x)}1 and ÿ1 are the linear accelerations of the front module, and {umlaut over (θ)}1 and {umlaut over (θ)}2 are the rotational accelerations of the first and second modules, respectively. The inertial linear velocities of a drive module are related to its forward velocity vxi in a body-centric frame:
{dot over (x)}i=vxi cos θ1
{dot over (y)}i=vxi sin θ1 (Eq. 5)
where i=1, 2, where 1 refers to the front module and 2 refers to the rear module. The body-centric forward velocity of module i is related to its left and right wheel rotational velocities:
where rw is the wheel radius, ωri is the right wheel rotational velocity and ωli is the left wheel rotational velocity. The rotational velocity of a drive module is also related to its wheel velocities:
where w is the distance between the right and left wheels of the drive module.
If the front and rear modules are to follow the same circular path on the plane, there will be a yaw joint angle θj,d that will permit the rear module to follow the front module. Consider the moment the front module enters a curve, while the rear module maintains its previous heading. The instantaneous center of rotation of the front module, rIC, is:
The path length s that must be traveled by the rear module before it enters the same curve is
s=rICθd=d1+d2 (Eq. 9)
At the moment the rear module enters the curve, the angle θd is the difference in orientation between the front and rear modules. This angle can be calculated from the instantaneous center of rotation by rewriting Equation (9):
Note that if the robot is traveling on a straight path, rIC is infinite and the desired joint angle is 0.
A simple leader-follower controller is desired so that the rear module follows the same path as the front module. The user directly drives the front module using a joystick, which sends forward and turning velocity commands that are converted into left and right motor speeds. The leader-follower controller must determine the commands to send to the right and left rear motors. For straight-line motion, one could simply match all the motor speeds, but this will not work while turning. There is a path-dependent yaw joint angle that permits the rear module to follow the front module regardless of the path curvature. Using the kinematic equations above, a feedback controller that drives the yaw joint angle to the desired angle can be implemented.
The robot is equipped with wheel encoders, an IMU on each module, and an angular position sensor on its yaw joint. The errors between the commanded and measured left and right wheel speeds of the front module are computed at block 1100. Errors between the commanded and measured left and right wheel speeds of the rear module are computed at block 1060. These errors are used by the front and rear module controllers, blocks 1090 and 1070, respectively, to send commands to the front and rear module motors (not shown). The torques applied by the robot motors cause robot motion at block 1080. The robot wheel speeds, accelerations and yaw joint angle are measured by the wheel encoders, IMUs and yaw joint encoder, respectively, at block 1110.
The error between the desired and measured joint angle is calculated as:
e=θj,m−θd (Eq. 11)
where θj,m is the measured joint angle. The change in this error can be calculated using the encoder measurements from both sets of wheels:
Using these errors, a PID control law can be defined:
u=Kpe+KI∫edt+KDė (Eq. 13)
where u is the commanded angular velocity of the rear module, and Kp, KI, and KD are the proportional, integral and derivative gains, respectively. The angular position error in Eq. 13 can be numerically integrated with sufficient anti-windup included in the software implementation. Both a forward velocity and angular velocity are needed to determine the right and left wheel commands. The angular velocity command is determined by the PID controller (
vx,2c=vx,1c (Eq. 14)
This is the kinematic constraint used at block 1050. The left and right rear motor speed commands, ω2,r,c and ω2,l,c, can be determined by:
The robot geometry imposes limits on its permissible trajectories. The yaw joint has a range of (−45/45) degrees, which restricts the robot's ability to maneuver in a tight circle. Hence, there is a minimum instantaneous center of rotation that limits the difference between the right and left motor speeds. This constraint is found by substituting the maximum range into Eq. 10 and rearranging:
A model of the system dynamics is needed to emulate its behavior when testing the leader-follower PID controller in simulation. Assume that the magnetic force between the wheels and the planar surface the robot travels on acts orthogonally to the robot motion. The following state model can be derived from first principles:
M(q){dot over (v)}+H(v,q)+P(q)v=Bu (Eq. 17)
where v is the state vector, M (q) is the mass matrix, vector H (v, q) contains centripetal terms, P (q) is the dissipative matrix, B is the input matrix, and u is the control vector. The state vector v is defined as
V=[ωr1ωl1ωr2ωl2ir1il1ir2il2]T (Eq. 18)
where i is the motor current. The system inputs are the voltages applied to the motors:
u=[Vr1Vl1Vr2Vl2]T (Eq. 19)
Operator control software may run on a portable device, such as a laptop or tablet, and provides the command and control interface for directing the system and receiving information back from the system. As an example, driver module software may run using Linux on a Gumstix® Overo Fire COM within each driver module. A wireless capability provided by the Overo Fire COM may be used for communication between driver modules and the OCU. Software on the system executes low-level tasks and communications such as battery-level reporting, motor control, and sensor-data routing. Motor control software provides basic movement commands from the OCU via the Overo processor to the motor controllers.
The OCU may comprise a portable device, such as a laptop or tablet, with the Microsoft® Windows operating system and multi-robot operator control unit (MOCU) application software. Used for controlling a wide variety of unmanned systems, MOCU is modular and scalable so it can be used with both existing and future platforms. The modularity has been extended to the user interface as well, making it possible to create the full gamut of user interfaces, ranging from headless to tiled windows to completely immersive game-like displays.
While the modules are used primarily for interfacing to different protocols, specialized hardware, video decoding and the like, most of the user interface is defined in Extensible Markup Language (XML) configuration files, making it relatively easy to customize what the display looks like and how the user interacts with the system, whether this be via mouse, keyboard, touchscreen, joystick, or other input devices. Control of the system through a MOCU will provide video feedback and joystick control for the system. In some embodiments, feedback from new sensors, battery life, robot pose, and other relevant information and control interfaces may be included.
Each Segment has its own processor and ability to send feedback and other information to other segments to coordinate mobility over an Ethernet link. Wireless and wired links may be used. As shown, master module 1310 is connected to slave modules 1320 and 1330 via an Ethernet link. Further, the power management systems of modules 1310, 1320, and 1330 may be connected via a power bus. Also shown, an OCU 1340 is connected to master segment 1350 and slave segments 1360 and 1370 via an Ethernet link.
Many modifications and variations of the Multi-Segmented Magnetic Robot are possible in light of the above description. Within the scope of the appended claims, the embodiments of the systems described herein may be practiced otherwise than as specifically described. The scope of the claims is not limited to the implementations and the embodiments disclosed herein, but extends to other implementations and embodiments as may be contemplated by those having ordinary skill in the art.
This application claims the benefit of U.S. Provisional Patent Application Ser. No. 62/069,684 filed Oct. 28, 2014, entitled “Multi-Segmented Climbing Robot”, the content of which is fully incorporated by reference herein.
Number | Name | Date | Kind |
---|---|---|---|
3437786 | Colinet | Apr 1969 | A |
3690393 | Guy | Sep 1972 | A |
4095378 | Urakami | Jun 1978 | A |
4977971 | Crane, III | Dec 1990 | A |
5161631 | Urakami | Nov 1992 | A |
5284096 | Pelrine | Feb 1994 | A |
5355807 | Pelrine | Oct 1994 | A |
5363935 | Schempf | Nov 1994 | A |
5551525 | Pack | Sep 1996 | A |
5752577 | Urakami | May 1998 | A |
5894901 | Awamura | Apr 1999 | A |
5921337 | Okamoto | Jul 1999 | A |
6000484 | Zoretich | Dec 1999 | A |
6125955 | Zoretich | Oct 2000 | A |
6267196 | Wilcox | Jul 2001 | B1 |
6691811 | Bruntrup | Feb 2004 | B2 |
6742617 | Jeswine | Jun 2004 | B2 |
6793026 | De Fazio | Sep 2004 | B1 |
7011171 | Poulter | Mar 2006 | B1 |
7233221 | Reboredo Losada | Jun 2007 | B2 |
8393422 | Pensel | Mar 2013 | B1 |
8464815 | Niederberger | Jun 2013 | B2 |
8540044 | Ba-abbad | Sep 2013 | B2 |
8935014 | Jacobsen | Jan 2015 | B2 |
20100139995 | Rudakevych | Jun 2010 | A1 |
20100263948 | Couture | Oct 2010 | A1 |
20110050374 | Dvorak | Mar 2011 | A1 |
20110240382 | Gettings | Oct 2011 | A1 |
20130289766 | Hafenrichter | Oct 2013 | A1 |
20140110021 | Divine | Apr 2014 | A1 |
20150019103 | Choi | Jan 2015 | A1 |
20150081092 | Jacobsen | Mar 2015 | A1 |
20150153170 | Gonzalez | Jun 2015 | A1 |
20150153312 | Gonzalez | Jun 2015 | A1 |
Number | Date | Country |
---|---|---|
102039592 | May 2012 | CN |
103009373 | Dec 2013 | CN |
02095989 | Apr 1990 | JP |
Entry |
---|
Tâche, F., Fischer, W., Siegwart, R., Moser, R. and Mondada, F., “Compact magnetic wheeled robot with high mobility for inspecting complex shaped pipe structures,” IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS) 2007, 261-266 (2007). |
Tâche, F., Fischer, W., Caprari, G., Siegwart, R., Moser, R. and Mondada, F., “Magnebike: A magnetic wheeled robot with high mobility for inspecting complex-shaped structures,” Journal of Field Robotics, 26, 453-476 (2009). |
Eich, M. and Vogele, T., “Design and control of a lightweight magnetic climbing robot for vessel inspection,” 19th Mediterranean Conf. on Control & Automation (MED), 1200-1205 (2011). |
Eiammanussakul, T., Taoprayoon, J. and Sangveraphunsiri, V., “Weld Bead Tracking Control of a Magnetic Wheel Wall Climbing Robot Using a Laser-Vision System,” Applied Mechanics and Materials, 219-223 (2014). |
Yi, Z., Gong, Y., Wang, Z. and Wang, X., “Development of a wall climbing robot for ship rust removal,” Int. Conf. on Mechatronics and Automation (ICMA), 4610-4615 (2009). |
Leon-Rodriguez, H., Hussain, S. and Sattar, T., “A compact wall-climbing and surface adaptation robot for nondestructive testing,” 12th Int. Conf. on Control, Automation and Systems (ICCAS), 404-409 (2012). |
Fernández, R., González, E., Feliú, V. and Rodríguez, A. G., “A wall climbing robot for tank inspection. An autonomous prototype,” in 36th Annual Conf. on IEEE Industrial Electronics Society, 1424-1429 (2010). |
Kitai, S., Tsuru, K. and Hirose, S., “The proposal of swarm type wall climbing robot system ‘Anchor Climber’—the design and examination of adhering mobile unit,” IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), 475-480 (2005). |
Silva, M. F., Barbosa, R. S. and Oliveira, A. L. C., “Climbing Robot for Ferromagnetic Surfaces with Dynamic Adjustment of the Adhesion System,” Journal of Robotics, 2012, (2012). |
Lee, G., Seo, K., Lee, S., Park, J., Kim, H., Kim, J. and Seo, T., “Compliant track-wheeled climbing robot with transitioning ability and high-payload capacity,” 2011 IEEE Int. Conf. on Robotics and Biomimetics (ROBIO), 2020-2024 (2011). |
Espinoza, R. V., de Oliveira, A. S., de Arruda, L. V. R. and Neves Junior, F., “Navigation's Stabilization System of a Magnetic Adherence-Based Climbing Robot,” Journal of Intelligent & Robotic Systems, 1-17 (2014). |
Lee, G., Wu, G., Kim, S. H., Kim, J. and Seo, T., “Combot: Compliant climbing robotic platform with transitioning capability and payload capacity,” 2012 IEEE Int. Conf. on Robotics and Automation (ICRA), 2737-2742 (2012). |
Wang, H., Huang, X., Hong, R. and Fang, C., “A new inspection robot system for storage tank,” 7th World Congress on Intelligent Control and Automation (WCICA), 7427-7431 (2008). |
Lee, W., “Proposition of reconfigurable wall climbing robot using 6DOF force torque sensor based on flexible structure for real environment,” 13th Int. Conf. on Control, Automation and Systems (ICCAS), 1802-1806 (2013). |
Erdmann, C. R., “080727-N-4236E-155,” United States Navy, <http://www.navy.mil/view—image.asp?id=62050> (2008). |
Burmeister, A., Pezeshkian, N., Talke, K., Ostovari, S., Everett, H.R., Hart, A., Gilbreath, G. and Nguyen, H.G., “Design of a Multi-Segmented Robot for Hull Inspection,” Naval Engineers Journal, 126-3, 53-60 (2014). |
Kamegawa, T., Yarnasaki, T., Igarashi, H. and Matsuno, F., “Development of the snake-like rescue robot ‘kohga’,” IEEE Int. Conf. on Robotics and Automation (ICRA '04), 5081-5086 (2004). |
Laumond, J. P., Sekhavat, S. and Lamiraux, F., “Guidelines in nonholonomic motion planning for mobile robots,” Robot Motion Planning and Control, 229, 1-53 (1998). |
Tanaka, M. and Matsuno, F., “Experimental Study of Redundant Snake Robot Based on Kinematic Model,” 2007 IEEE Int. Conf. on Robotics and Automation, 2990-2995 (2007). |
Michatek, M. M., “A highly scalable path-following controller for N-trailers with off-axle hitching,” Control Engineering Practice, 29, 61-73 (2014). |
Matsuno, F. and Suenaga, K., “Control of redundant snake robot based on kinematic model,” Proc. 41st SICE Annual Conference, 3, 1481-1486 (2002). |
Tilbury, D. Laumond, J. P., Murray, R., Sastry, S. and Walsh, G., “Steering car-like systems with trailers using sinusoids,” 1992 IEEE Int. Conf. on Robotics and Automation, 3, 1993-1998 (1992). |
Laumond, J. P., “Controllability of a multibody mobile robot,” IEEE Transactions on Robotics and Automation, 9, 755-763 (1993). |
Lamiraux, F., Sekhavat, S. and Laumond, J. P., “Motion planning and control for Hilare pulling a trailer,” IEEE Transactions on Robotics and Automation, 15, 640-652 (1999). |
Lee, J.-H., Chung, W., Kim, M. and Song, J.-B., “A passive multiple trailer system with off-axle hitching,” Int. Journal of Control Automation and Systems, 2, 289-297 (2004). |
Nakamura, Y., Ezaki, H., Tan, Y. and Chung, W., “Design of steering mechanism and control of nonholonomic trailer systems,” IEEE Transactions on Robotics and Automation, 17, 367-374 (2001). |
Sampei, M., Tamura, T., Itoh, T. and Nakamichi, M., “Path tracking control of trailer-like mobile robot,” Proc. IEEE/RSJ Int. Workshop on Intelligent Robots and Systems: Intelligence for Mechanical Systems (IROS '91), 1, 193-198 (1991). |
Yuan, J., Huang, Y., Kang, Y. and Liu, Z., “A Strategy of Path Following Control for Multi-Steering Tractor-Trailer Mobile Robot,” IEEE Int. Conf. on Robotics and Biomimetics (ROBIO 2004), 163-168 (2004). |
Kerber, M., et al., “Design of a Multi-Segmented Robot for Hull Climbing”, SPAWAR Systems Center Pacific (SSC Pacific) Technical report 2028, San Diego, CA, Nov. 2013. |
Kerber, M., et al., “Design and Optimization of a Magnetic Wheel for Hull Climbing”, SPAWAR Systems Center Pacific (SSC Pacific) Technical report 2025, San Diego, CA, Nov. 2013. |
Magnetic wheels, available at http://www.davehylands.com/Robotics/Wedgy/07-Magnetic-Wheel.html. |
Gecko—Magnetic Wall Climbing Robot, available at http://www.instructables.com/id/Gecko-Magnetic-wall-climbing-robot/. |
Number | Date | Country | |
---|---|---|---|
62069684 | Oct 2014 | US |