Formation Control of Wheeled Robots in the Port-Hamiltonian Framework

Vos, Ewoud; Scherpen, Jacquelien M.A.; Schaft, A.J. van der; Postma, Ate

doi:10.3182/20140824-6-za-1003.00394

Cited by 22 publications

(35 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we present the results of the stability and convergence analysis of the closed loop system . Based on , a network of strictly output passive nonholonomic wheeled robots in the absence of disturbances converges to the desired formation where agents' momenta are zero. Here, we first consider a network of passive agents such that at least one of the agents is strictly output passive and study the closed‐loop system considering harmonic and constant disturbances.…”

Section: Results and Analysismentioning

confidence: 99%

“…The output

{\overset{y}{true}}_{i}

is transformed accordingly into the new output y i =( y x , i , y y , i ). The transformation

(ū_{i}, {\overset{y}{true}}_{i}) \mapsto (u_{i}, y_{i})

is defined as

ū_{i} = G_{i} (q_{i}) u_{i}, y_{i} = G_{i}^{T} (q_{i}) {\overset{y}{true}}_{i}

, where

G_{i} (q_{i}) = ((), \begin{matrix} \cos ϕ_{i} & \sin ϕ_{i} \\ - d_{AC, i} \sin ϕ_{i} & d_{AC, i} \cos ϕ_{i} \end{matrix}) .

The dynamics of robot i with new input u i and new output y i follows by substituting

ū_{i} = G_{i} (q_{i}) u_{i}, y_{i} = G_{i}^{T} (q_{i}) {\overset{y}{true}}_{i}

into .…”

Section: Problem Statementmentioning

confidence: 99%

“…We consider the control law u i as the sum of two control laws

u_{i}^{f}

and

u_{i}^{d}

. Hence, we can rewrite as follows:

\begin{matrix} (\begin{matrix} {\overset{q}{true}}_{i} \\ {\overset{p}{true}}_{i} \end{matrix}) & = (\begin{matrix} 0 & S_{i} (q_{i}) \\ - S_{i}^{T} (q_{i}) & - D_{i}^{r} \end{matrix}) (\begin{matrix} \frac{\partial H_{i}^{r}}{\partial q_{i}} \\ \frac{\partial H_{i}^{r}}{\partial p_{i}} \end{matrix}) + (\begin{matrix} 0 \\ G_{i} (q_{i}) \end{matrix}) (u_{i}^{f} + u_{i}^{d} + d_{i}) \\ y_{i} & = G_{i}^{T} (q) \frac{\partial H_{i}^{r}}{\partial p_{i}} . \end{matrix}

The control law

u_{i}^{f}

achieves the desired formation, based on the results by . The control law

u_{i}^{d}

is designed based on an internal‐model‐based approach (e.g., ) to reject the matched input disturbance d i in .…”

Section: Problem Statementmentioning

confidence: 99%

“…This assumption simplifies the problem; however, it is of high practical interests due to numerous applications, that is, controlling the gripper position located on the front end of a wheeled robot. A similar approach in the port-Hamiltonian framework and using passivity-based design techniques has been considered in [21].Main contribution. This work analyzes the formation control of a network of nonholonomic wheeled robots to achieve a desired formation despite of matched input disturbances.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Disturbance rejection in formation keeping control of nonholonomic wheeled robots

Jafarian

Vos

Persis

et al. 2016

Int. J. Robust. Nonlinear Control

Self Cite

View full text Add to dashboard Cite

SUMMARYThis paper presents the results of formation keeping control of a group of nonholonomic wheeled robots within the port-Hamiltonian framework and in the presence of matched input disturbances. Two scenarios on the internal damping of the dynamics of the robots are considered: strictly output passive and loss less robots. For strictly output passive robots, the distributed formation keeping controllers drive the robots towards a desired formation, while internal-model-based controllers locally compensate the harmonic input disturbance for each of the robots. Moreover, the effect of constant input disturbances is studied considering internal-model-based controllers. For lossless robots, results on formation keeping control are presented. Simulation results illustrate the effectiveness of the approach.

show abstract

Section: Results and Analysismentioning

confidence: 99%

“…The output

{\overset{y}{true}}_{i}

is transformed accordingly into the new output y i =( y x , i , y y , i ). The transformation

(ū_{i}, {\overset{y}{true}}_{i}) \mapsto (u_{i}, y_{i})

is defined as

ū_{i} = G_{i} (q_{i}) u_{i}, y_{i} = G_{i}^{T} (q_{i}) {\overset{y}{true}}_{i}

, where

G_{i} (q_{i}) = ((), \begin{matrix} \cos ϕ_{i} & \sin ϕ_{i} \\ - d_{AC, i} \sin ϕ_{i} & d_{AC, i} \cos ϕ_{i} \end{matrix}) .

The dynamics of robot i with new input u i and new output y i follows by substituting

ū_{i} = G_{i} (q_{i}) u_{i}, y_{i} = G_{i}^{T} (q_{i}) {\overset{y}{true}}_{i}

into .…”

Section: Problem Statementmentioning

confidence: 99%

“…We consider the control law u i as the sum of two control laws

u_{i}^{f}

and

u_{i}^{d}

. Hence, we can rewrite as follows:

\begin{matrix} (\begin{matrix} {\overset{q}{true}}_{i} \\ {\overset{p}{true}}_{i} \end{matrix}) & = (\begin{matrix} 0 & S_{i} (q_{i}) \\ - S_{i}^{T} (q_{i}) & - D_{i}^{r} \end{matrix}) (\begin{matrix} \frac{\partial H_{i}^{r}}{\partial q_{i}} \\ \frac{\partial H_{i}^{r}}{\partial p_{i}} \end{matrix}) + (\begin{matrix} 0 \\ G_{i} (q_{i}) \end{matrix}) (u_{i}^{f} + u_{i}^{d} + d_{i}) \\ y_{i} & = G_{i}^{T} (q) \frac{\partial H_{i}^{r}}{\partial p_{i}} . \end{matrix}

The control law

u_{i}^{f}

achieves the desired formation, based on the results by . The control law

u_{i}^{d}

is designed based on an internal‐model‐based approach (e.g., ) to reject the matched input disturbance d i in .…”

Section: Problem Statementmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Disturbance rejection in formation keeping control of nonholonomic wheeled robots

Jafarian

Vos

Persis

et al. 2016

Int. J. Robust. Nonlinear Control

Self Cite

View full text Add to dashboard Cite

show abstract

“…The final goal is a collaborating 'team' of air, ground and water-based robots. The ROSE project is divided in two parts; the first part concerns the design of algorithms for coordinating the network of robotic sensors, was carried out at the University of Groningen and is reported in (Vos, 2015). The second part is the design of the energy-autonomous vehicle which is the focus of this work.…”

Section: Introductionmentioning

confidence: 99%

Controlled passive actuation : concepts for energy efficient actuation using mechanical storage elements and continuously variable transmissions

Dresscher¹

View full text Add to dashboard Cite

SummaryEnergy autonomous robots need to take care of their own energy supply. In the absence of a charging station, the robot will have to replenish its energy in other ways. Alternative energy sources available to a robot in an outdoor environment may include unrefined biomass, sun and wind. As these energy sources are limited, energy efficiency of mobile robots is of key importance.With this research, we want to make a step toward energy autonomous legged robots. For this, we have two key objectives: increase insight in the mechanisms of energy loss and contribute toward the improvement of energy efficiency.In the first part of this thesis, the energy losses in legged robots are discussed. The active supporting of the system mass and the additional movement of the legs contribute to the high energy consumption in legged vehicles. However, how significant the contributions of the various causes of energy loss are, is not well understood. We investigate the mechanisms of energy loss in electric actuators to understand their contribution to energy (in)efficiency. As a part of this effort, a method to select a motor-gearbox combination for energy efficiency is presented and evaluated to show the possible impact. We conclude that for most robotic applications, electric actuators are the dominant cause of energy losses. Selecting a motor-gearbox combination for energy efficiency can provide a significant improvement in the efficiency. However, the energy losses in the motor-gearbox combination remain significant.As an alternative to using only electric actuators for actuation, an actuation principle that incorporates a mechanical storage element and a Continuously Variable Transmission (CVT) in the drive-train is discussed. This actuation principle is named Controlled Passive Actuation (CPA). The envisioned benefits are twofold: first, separation of the energy supply function and the servoing function enables using the electric actuator in its most efficient range of operation; and, second, by incorporating the capability to recycle energy within the mechanical domain, the energy loss in the electric actuator can be avoided.An important component in this actuation principle is the CVT and in the ii second part of this thesis, two CVT's that were designed for application in Controlled Passive Actuation are discussed. The first design in a lever with a pivot that moves along the lever. It is shown that the design is conceptually suitable for application in Controlled Passive Actuation and that the mechanical design needs improvement. The second design is based on two spheres of which the relative orientation can be changed to adapt the transmission ratio. This concept is named Dual-Hemi CVT. A prototype was built and experimental results show that it is suitable for application in CPA.In the third part of this thesis, two versions of CPA are discussed. In the first version of CPA, a flywheel is used as mechanical storage element. By changing the rate of the transmission ratio of the Continuously Variable Transmission,...

show abstract

Port-Hamiltonian Control of a Differential Robot

Benavides

García

2021

Advances in Automation and Robotics Research

View full text Add to dashboard Cite

A novel approach to tracking control of a differential robot is proposed. Based on a Port-Hamiltonian representation of a differential robot, canonical transformations are used to obtain a controller that guarantees the tracking of a reference. The control law is inspired by a well-known controller taken from the literature and the Port-Hamiltonian version of the control law is derived. The performance of the algorithm as well as some of the robustness properties are tested using simulations under different scenarios.

show abstract

Formation Control of Wheeled Robots in the Port-Hamiltonian Framework

Cited by 22 publications

References 7 publications

Disturbance rejection in formation keeping control of nonholonomic wheeled robots

Disturbance rejection in formation keeping control of nonholonomic wheeled robots

Controlled passive actuation : concepts for energy efficient actuation using mechanical storage elements and continuously variable transmissions

Port-Hamiltonian Control of a Differential Robot

Contact Info

Product

Resources

About