Abstract
This study presents a novel mechanism for the control of a wheeled robot maneuvering with a single actuator. Elements of snakeboard and two-wheeled skateboard propulsion are applied to the design. Two passive wheels, i.e., casters whose orientation can be controlled, are attached in the back and front of the robot body, and a rotor rotates above the body to induce body propulsion using its counter torque. Three degrees of freedom of motion, i.e., the orientation of the rotor and each of the two casters, are mechanically coupled to the single actuator via a torque limiter. The stoppers are set to restrict the angle of the caster orientation, and a torque limiter allows the rotor to continue rotating without being affected by the stopper’s restriction to the range of motion. Experiments demonstrate that the sinusoidal rotor rotation can propel this robot forward and that adding the increasing or decreasing offset to the sinusoidal rotor rotation can curve the robot’s motion. Next, a method to position the robot at a specified goal position is proposed, assuming that the current position of the robot is detectable in every control cycle. This method adjusts the rate of increase or decrease of the offset in sinusoidal rotor rotation depending on the direction of the goal position. Introducing the motion capture system enables the robot to successfully reach the specified goal positions.
Similar content being viewed by others
References
Kajita, S., Kanehiro, F., Kaneko, K., Yokoi, K., Hirukawa, H.: The 3d linear inverted pendulum mode: A simple modeling for a biped walking pattern generation. In: Proceedings. 2001 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2001, vol. 1, pp. 239–246 IEEE (2001)
Goswami, A., Thuilot, B., Espiau, B.: A study of the passive gait of a compass-like biped robot Symmetry and chaos. Int. J. Robot. Res. 17(12), 1282–1301 (1998)
Hirose, S.: Three basic types of locomotion in mobile robots. In: Fifth International Conference on Advanced Robotics ’Robots in Unstructured Environments’, pp. 12–17. IEEE (1991)
Hirose, S., Yamada, H.: Snake-like robots [tutorial]. IEEE Robot. Autom. Mag. 16(1), 88–98 (2009)
Takemori, T., Tanaka, M., Matsuno, F.: Gait design for a snake robot by connecting curve segments and experimental demonstration. IEEE Transactions on Robotics (2018)
Ariizumi, R., Tanaka, M., Matsuno, F.: Analysis and heading control of continuum planar snake robot based on kinematics and a general solution thereof. Adv. Robot. 30(5), 301–314 (2016)
Krishnaprasad, PS, Tsakiris, D.P.: Oscillations, Se (2)-snakes and motion control: A study of the roller racer. Dynamical Systems: An International Journal 16(4), 347–397 (2001)
Mori, H., Nagamine, T., Ichuo, T., Sato, Y.: Effects of friction on driving mechanism using swing motion. Transactions of the JSME (in Japanese) 81(832), 15–00330 (2015)
Iannitti, S., Lynch, K.M.: Minimum control-switch motions for the snakeboard: A case study in kinematically controllable underactuated systems. IEEE Trans. Robot. 20(6), 994–1006 (2004)
Ostrowski, J.: Steering for a class of dynamic nonholonomic systems. IEEE Trans. Autom. Control 45(8), 1492–1498 (2000)
Narikiyo, T.: Control of underactuated mechanical systems via passive velocity field control: Application to snakeboard and 3D rigid body. Nonlinear Anal. 71, e2358–e2365 (2009)
Shammas, E., De Oliveira, M.: Motion planning for the snakeboard. Int. J. Robot. Res. 31(7), 872–885 (2012)
Derammelaere, S., Copot, C., Haemers, M., Verbelen, F., Vervisch, B., Ionescu, C., Stockman, K.: Realtime locomotion control of a snakeboard robot based on a novel model, enabling better physical insights. European Journal of Control (2018)
Salman, H., Dear, T., Babikian, S., Shammas, E., Choset, H.: A Physical parameter-based skidding model for the snakeboard. In: 2016 IEEE 55Th Conference on Decision and Control (CDC ), pp. 7555–7560 (2016)
Wang, T., Su, B., Kuang, S., Wang, J.: On kinematic mechanism of a two-wheel skateboard: The essboard. J. Mech. Robot. 5(3), 034503 (2013)
Su, B., Wang, T., Wu, R., Wang, J.: Infimum of path length of nonholonomic vehicle with finitely bounded curvature radius. J. Intell. Robot. Syst. 79(2), 197–210 (2015)
Kinugasa, K., Ishikawa, M., Sugimoto, Y., Osuka, K.: Modeling and control of casterboard robot. IFAC Proceedings Volumes 46(23), 785–790 (2013)
Ito, S., Sugiura, S., Masuda, Y., Kiely, S., Yabuki, J., Morita, R.: A mechanism of single actuator snakeboard robot and its curving motion generation. In: 2018 IEEE International Conference on Robotics and Biomimetics (IEEE ROBIO 2018), pp. 1232–1237 (2018)
Bullo, F., Lewis, A.D.: Kinematic controllability and motion planning for the snakeboard. IEEE Trans. Robot. Autom. 19(3), 494–498 (2003)
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interests
The authors declare that they have no conflict of interest.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
Appendix
Appendix
1.1 A1 Motion Equation
The position and the orientation of the man body, the front and the rear wheel are denoted by (X0, Y0, ϕ0), (X1, Y1, ϕ1) and (X2, Y2, ϕ2), respectively. The positional relationship among them is represented as follows:
Here, L1 and L2 are the distance from the center of the main body to each wheel. The velocity of the front and rear wheel normal to the wheel axis,V1 and V2, are given as follows:
while the constraints that keep the wheels from slipping to each wheel axis direction are written as follows:
where, i = 1, 2 distinguishes the front and rear wheels.
The rotor, the torque limiter and the wheels move together with the main body. The main body dynamics is obtained as the following motion equation:
Here, \(\mathrm {Q} = \left [\begin {array}{ccc} X_{0} & Y_{0} & \phi _{0} \end {array} \right ]^{T}\) is the state vector of the main body, M is the inertial matrix whose components are given by the total mass of the main body including the rotor and wheels M0, and the moment of inertial of the main bodyI0 (without rotor),
JC and JV are the Jacobian matrices
that relates \(\dot {Q}\) to
respectively, and \(\mathrm {F}_{C}= \left [\begin {array}{cc} F_{C1} & F_{C2} \end {array} \right ]^{T}\) is the constraint force that prevents each wheel from slipping to the wheel axis direction, \(\mathrm {F}_{V} = \left [\begin {array}{cc} F_{V1} & F_{V2} \end {array} \right ]^{T}\) is the resistance force against the wheel rotation or progression such as the friction. Ff is the friction force, given by the following equation:
See the Section A for θR and ξR. JU becomes as follws:
And, u is control input, i.e., the counter force against the rotor rotation u = −τ.
1.2 A2 Velocity Constraints
The wheel is assumed not to slip in its wheel-axis direction. This condition is equivalent to V⊥ == 0, which is written using the derivative of (11)–(14) as follows:
1.3 A3 Resistance Force
As the resistance force applied to the wheel motion, the viscous friction is considered. Then, FV is represented as follows:
where, B is the viscous matrix:
b1 and b2 are the viscous coefficients corresponding to each wheel.
1.4 A4 Stopper
The collision and contact between the bar and stoppers are modeled as the spring and the dumper, which generates the reaction torque τstopper around the wheel rotation.
1.5 A5 Coupling Dynamics of Wheel Orientations
Put the orientation angle of the front and the rear wheel relative to the main body as θ1 = ϕ1 −ϕ0 and θ2 = ϕ2 −ϕ0, respectively, and let the torque limiter angle relative to the main body θ3. Their dynamics are given as follows:
Here, ξ1, ξ2, ξ3 are the viscous coefficients of the rotation around the front wheel, the rear wheel, and the rotor, and τ1, τ2 and τ3 are the torque around them, respectively. The coupling of the wheel orientation imposes the following constraints:
Besides, the stopper limits the range of the rotation, and the torque generated by the motor is transmitted via the torque limiter. Considering all of them, the rotation of the wheel orientation is solely represented by the following dynamics of θ because of the coupling (33)
1.6 A6 Rotor
The motion of the rotor is simply modeled as the rotation of the rigid bar.
Here θR is the rotor angle relative to the main body, IR is the moment of inertia of the rotor, ξR is the friction coefficient of the rotor rotation, and τR denotes the driving moment.
1.7 A7 Actual Computation
The degrees of freedom of this system becomes five, i.e., X0, Y0, ϕ0, θ and θR. The state of the main body is represented by X0, Y0 and ϕ0, whose dynamics are given by (17) with the velocity constraint (16), and is driven as the torque −τ.
Then, its reaction forceτ rotates the rotor as well as the wheel orientation. Namely,
However, the dynamics of the rotor and the wheel orientation varies depending on whether the torque limiter is on or off.
If the torque limiter is off, the wheel orientation changes with the rotor rotation and the coupling relation
holds. Accordingly, (34) and (35) should be solved under this condition which allows us to get \(\ddot {\theta } (= \ddot \theta _{R})\). Now, we can obtain the value of the τlimiter(τw) from this \(\ddot {\theta }\), using (34). Here,
- a.
The case τmin ≤ τlimiter (τw) ≤ τmax: The torque limiter is actually off. Thus, we should continue the calculation without any modification.
- b.
The case τlimiter (τw) < τmin: The torque limiter should disconnect the torque transmission. Thus, put τw = τmin and calculate (34) and (35) again removing condition (37).
- c.
The case τlimiter (τw) > τmax: The torque limiter should disconnect the torque transmission. Thus, put τw = τmax and calculate (34) and (35) again removing condition (27).
Finally, we should mention the calculation for the moment at which the torque limiter switches from on to off. In this moment, the two different rotations, \(\dot \theta \) and \(\dot \theta _{R}\), are coupled and starts to rotate with the same speed \(\dot \theta _{new}\). This \(\dot \theta _{new}\) is calculated considering the conservation of the angular moment:
Namely, the rotation speed jumps at this moment.
1.8 A8 Default Parameters
The Rung-Kutta method with 0.0001 second step size was utilized for the simulations. The followings are default parameter values in the simulations when any other descriptions are now shown: M = 4, I0 = 0.08, I1 = I2 = 0.002, I3 = 0.001, IR = 0.05, L1 = L2 = 0.1, γ1 = 1, γ2 = − 1, γ = 1, ξ0 = 0.01, ξ1 = ξ2 = ξ = 0, ξR = 0.1, b1 = b2 = − 3, bs = 0.2, ks = 5, θL+ = +α, θL− = −α. τmin = − 0.1, τmax = 0.1,
Rights and permissions
About this article
Cite this article
Ito, S., Sugiura, S., Masuda, Y. et al. Mechanism and Control of a One-Actuator Mobile Robot Incorporating a Torque Limiter. J Intell Robot Syst 97, 431–448 (2020). https://doi.org/10.1007/s10846-019-01036-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10846-019-01036-8