Skip to main content
Log in

Controlled Lagrangians and stabilization of Euler–Poincaré mechanical systems with broken symmetry II: potential shaping

  • Original Article
  • Published:
Mathematics of Control, Signals, and Systems Aims and scope Submit manuscript

Abstract

We apply the method of controlled Lagrangians by potential shaping to Euler–Poincaré mechanical systems with broken symmetry. We assume that the configuration space is a general semidirect product Lie group \({\mathsf {G}} \ltimes V\) with a particular interest in those systems whose configuration space is the special Euclidean group \(\mathsf {SE}(3) = \mathsf {SO}(3) \ltimes {\mathbb {R}}^{3}\). The key idea behind the work is the use of representations of \({\mathsf {G}} \ltimes V\) and their associated advected parameters. Specifically, we derive matching conditions for the modified potential exploiting the representations and advected parameters. Our motivating examples are a heavy top spinning on a movable base and an underwater vehicle with non-coincident centers of gravity and buoyancy. We consider a few different control problems for these systems, and show that our results give a general framework that reproduces our previous work on the former example and also those of Leonard on the latter. Also, in one of the latter cases, we demonstrate the advantage of our representation-based approach by giving a simpler and more succinct formulation of the problem.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Similar content being viewed by others

References

  1. Blankenstein G, Ortega R, van der Schaft AJ (2002) The matching conditions of controlled Lagrangians and IDA-passivity based control. Int J Control 75(9):645–665

    Article  MathSciNet  Google Scholar 

  2. Bloch AM, Leonard NE, Marsden JE (1999) Potential shaping and the method of controlled lagrangians. In: Proceedings of the 38th IEEE conference on decision and control (Cat. No.99CH36304), vol. 2, pp 1652–1657

  3. Bloch AM, Leonard NE, Marsden JE (2001) Controlled Lagrangians and the stabilization of Euler–Poincaré mechanical systems. Int J Robust Nonlinear Control 11(3):191–214

    Article  Google Scholar 

  4. Bloch AM, Leonard NE, Marsden JE (2000) Controlled Lagrangians and the stabilization of mechanical systems. I. The first matching theorem. IEEE Trans Autom Control 45(12):2253–2270

    Article  MathSciNet  Google Scholar 

  5. Bloch AM, Chang DE, Leonard NE, Marsden JE (2001) Controlled Lagrangians and the stabilization of mechanical systems. II. Potential shaping. IEEE Trans Autom Control 46(10):1556–1571

    Article  MathSciNet  Google Scholar 

  6. Borum AD, Bretl T (2014) Geometric optimal control for symmetry breaking cost functions. In: 53rd IEEE conference on decision and control, pp 5855–5861

  7. Borum AD, Bretl T (2016) Reduction of sufficient conditions for optimal control problems with subgroup symmetry. IEEE Trans Autom Control 99:3209–3224 (ISSN 0018-9286)

    MathSciNet  MATH  Google Scholar 

  8. Bullo F, Lewis AD (2004) Geometric control of mechanical systems, volume 49 of texts in applied mathematics. Springer

  9. Cendra H, Holm DD, Marsden JE, Ratiu TS (1998) Lagrangian reduction, the Euler–Poincaré equations, and semidirect products. Amer Math Soc Trans 186:1–25

    MATH  Google Scholar 

  10. Chang DE, Marsden JE (2004) Reduction of controlled Lagrangian and Hamiltonian systems with symmetry. SIAM J Control Optim 43(1):277–300

    Article  Google Scholar 

  11. Chang DE, Bloch AM, Leonard NE, Marsden JE, Woolsey CA (2002) The equivalence of controlled Lagrangian and controlled Hamiltonian systems. ESAIM: COCV 8:393–422

    MathSciNet  MATH  Google Scholar 

  12. Chyba M, Haberkorn T, Smith RN, Wilkens GR (2007) Controlling a submerged rigid body: a geometric analysis. In: Bullo F, Fujimoto K (eds), Lagrangian and Hamiltonian methods for nonlinear control 2006. Springer, Berlin, pp 375–385

  13. Contreras C, Ohsawa T (2020) Controlled Lagrangians and stabilization of Euler–Poincaré mechanical systems with broken symmetry I: Kinetic shaping

  14. Hamberg J (1999) General matching conditions in the theory of controlled Lagrangians. In: Proceedings of the 38th IEEE conference on decision and control, 1999, vol 3, pp 2519–2523

  15. Hamberg J (2000) Controlled Lagrangians, symmetries and conditions for strong matching. In IFAC Lagrangian and Hamiltonian methods for nonlinear control

  16. Holm DD, Marsden JE, Ratiu TS (1998) The Euler–Poincaré equations and semidirect products with applications to continuum theories. Adv Math 137(1):1–81

    Article  MathSciNet  Google Scholar 

  17. Holm DD, Schmah T, Stoica C (2009) Geometric mechanics and symmetry: from finite to infinite dimensions. Oxford texts in applied and engineering mathematics. Oxford University Press

  18. Leonard NE (1997a) Stability of a bottom-heavy underwater vehicle. Automatica 33(3):331–346

    Article  MathSciNet  Google Scholar 

  19. Leonard NE (1997b) Stabilization of underwater vehicle dynamics with symmetry-breaking potentials. Syst Control Lett 32(1):35–42

    Article  MathSciNet  Google Scholar 

  20. Leonard NE, Marsden JE (1997) Stability and drift of underwater vehicle dynamics: mechanical systems with rigid motion symmetry. Physica D 105(1–3):130–162

    Article  MathSciNet  Google Scholar 

  21. Marsden JE, Ratiu TS (1999) Introduction to mechanics and symmetry. Springer, New York

    Book  Google Scholar 

  22. Marsden JE, Ratiu TS, Weinstein A (1984a) Semidirect products and reduction in mechanics. Trans Am Math Soc 281(1):147–177

    Article  MathSciNet  Google Scholar 

  23. Marsden JE, Ratiu TS, Weinstein A (1984) Reduction and Hamiltonian structures on duals of semidirect product Lie algebras. In Fluids and plasmas: geometry and dynamics, volume 28 of Contemporary Mathematics. American Mathematical Society

  24. Nijmeijer H, van der Schaft A (1990) Nonlinear dynamical control systems. Springer, New York

    Book  Google Scholar 

  25. Ortega R, Perez J, Nicklasson P, Sira-Ramirez H (1998) Passivity-based Control of Euler–Lagrange systems: mechanical. Electrical and electromechanical applications. Communications and control engineering. Springer, London

  26. Ortega R, van der Schaft AJ, Mareels I, Maschke B (2001) Putting energy back in control. IEEE Control Syst 21(2):18–33

    Article  Google Scholar 

  27. Ortega R, Spong MW, Gomez-Estern F, Blankenstein G (2002) Stabilization of a class of underactuated mechanical systems via interconnection and damping assignment. IEEE Trans Autom Control 47(8):1218–1233

    Article  MathSciNet  Google Scholar 

  28. Smith RN, Chyba M, Wilkens GR, Catone CJ (2009) A geometrical approach to the motion planning problem for a submerged rigid body. Int J Control 82(9):1641–1656

    Article  MathSciNet  Google Scholar 

  29. Spong MW, Bullo F (2005) Controlled symmetries and passive walking. IEEE Trans Autom Control 50(7):1025–1031

    Article  MathSciNet  Google Scholar 

  30. van der Schaft AJ (1986) Stabilization of Hamiltonian systems. Nonlinear Anal Theory Methods Appl 10(10):1021–1035

    Article  MathSciNet  Google Scholar 

  31. Woolsey CA, Leonard NE (2002) Stabilizing underwater vehicle motion using internal rotors. Automatica 38(12):2053–2062

    Article  MathSciNet  Google Scholar 

  32. Woolsey CA, Techy L (2009) Cross-track control of a slender, underactuated AUV using potential shaping. Ocean Eng 36(1):82–91

    Article  Google Scholar 

Download references

Acknowledgements

We would like to thank the reviewers for their helpful comments.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Tomoki Ohsawa.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This work was supported by NSF grant CMMI-1824798.

Lie–Poisson brackets

Lie–Poisson brackets

While this paper focuses on the Lagrangian formulation of mechanical systems with broken symmetry, one can perform the Legendre transformation to obtain the Hamiltonian formulation of the systems as well. The main advantage of the Hamiltonian formulation is that it is more useful in finding the Casimirs.

1.1 Lie–Poisson bracket on \({\mathfrak {s}}^{*} = ({\mathfrak {g}} \ltimes V)^{*}\)

Let \({\mathfrak {s}} = {\mathfrak {g}} \ltimes V\) be the Lie algebra of the semidirect product Lie group \({\mathsf {S}} \mathrel {\mathop :}={\mathsf {G}} \ltimes V\). The \((-)\)-Lie–Poisson bracket on \({\mathfrak {s}}^{*}\) is given by (see Marsden et al. [22, 23])

$$\begin{aligned} \left\{ f,h\right\} _{{\mathfrak {s}}^{*}}(\mu ,p) = -{\left\langle \mu , \left[ \frac{\delta f}{\delta \mu }, \frac{\delta h}{\delta \mu } \right] \right\rangle } -{\left\langle p, \rho '{\left( \frac{\delta f}{\delta \mu }\right) } \frac{\partial h}{\partial p} - \rho '{\left( \frac{\delta h}{\delta \mu }\right) } \frac{\partial f}{\partial p} \right\rangle } \end{aligned}$$
(31)

We denote \({\mathfrak {s}}^{*}\) equipped with \(\left\{ \,\cdot \,,\,\cdot \,\right\} _{{\mathfrak {s}}^{*}}\) by \({\mathfrak {s}}^{*}\).

Example 10

(Lie–Poisson bracket on \(\mathfrak {se}(3)^{*}\)) If \({\mathfrak {g}} = \mathfrak {so}(3)\) and \(V = {\mathbb {R}}^{3}\), then \({\mathfrak {s}} = \mathfrak {se}(3)\). Using the expression for \(\rho '\) from (4), (31) yields

$$\begin{aligned} \left\{ f,h\right\} _{\mathfrak {se}(3)^{*}}(\varvec{\Pi },{\mathbf {P}}) = -\varvec{\Pi }\cdot {\left( \frac{\partial f}{\partial \varvec{\Pi }} \times \frac{\partial h}{\partial \varvec{\Pi }} \right) } - {\mathbf {P}}\cdot {\left( \frac{\partial f}{\partial \varvec{\Pi }} \times \frac{\partial h}{\partial {\mathbf {P}}} - \frac{\partial h}{\partial \varvec{\Pi }} \times \frac{\partial f}{\partial {\mathbf {P}}} \right) }.\quad \end{aligned}$$
(32)

This is essentially the heavy top bracket upon replacing \({\mathbf {P}}\) by \(\varvec{\Gamma }\). In our context, \({\mathbf {P}}\) stands for the linear impulse defined in (9), and so has a different physical meaning from \(\varvec{\Gamma }\).

1.2 Lie–Poisson bracket on \(({\mathfrak {s}} \ltimes X)^{*}\)

We may describe those uncontrolled mechanical systems with broken symmetry shown in Sect. 3.3 as the Lie–Poisson equation on the dual \(({\mathfrak {s}} \ltimes X)^{*}\) of the semidirect product Lie algebra \({\mathfrak {s}} \ltimes X\). Particularly, using the representation \(\sigma \) defined in Sect. 3.2, the Lie–Poisson bracket on \(({\mathfrak {s}} \ltimes X)^{*}\) is given by

$$\begin{aligned} \left\{ f,h\right\} _{({\mathfrak {s}} \ltimes X)^{*}}(\mu ,p,a) = \left\{ f,h\right\} _{{\mathfrak {s}}^{*}} -{\left\langle a, \sigma '{\left( \frac{\delta f}{\delta (\mu ,p)}\right) } \frac{\partial h}{\partial a} - \sigma '{\left( \frac{\delta h}{\delta (\mu ,p)}\right) } \frac{\partial f}{\partial a} \right\rangle }. \end{aligned}$$
(33)

Also, by considering a subrepresentation on \(({\mathfrak {s}} \ltimes X)^{*}\), the controlled system (21) with potential shaping using the matching described in Sect. 4.2 may also be described in terms of the Lie–Poisson bracket on \(({\mathfrak {s}} \ltimes {\tilde{X}})^{*}\).

Example 11

(Lie–Poisson bracket on \((\mathfrak {se}(3) \ltimes {\mathbb {R}}^{3})^{*}\)) If \({\mathfrak {s}} = \mathfrak {se}(3)\) and \(X = {\mathbb {R}}^{3}\), then, using the bracket (32) and also the expression for \(\sigma '\) from (10), (33) gives

$$\begin{aligned} \left\{ f,h\right\} _{(\mathfrak {se}(3) \ltimes {\mathbb {R}}^{3})^{*}}(\varvec{\Pi },{\mathbf {P}},\varvec{\Gamma })&= -\varvec{\Pi }\cdot {\left( \frac{\partial f}{\partial \varvec{\Pi }} \times \frac{\partial h}{\partial \varvec{\Pi }} \right) } - {\mathbf {P}}\cdot {\left( \frac{\partial f}{\partial \varvec{\Pi }} \times \frac{\partial h}{\partial {\mathbf {P}}} \!-\! \frac{\partial h}{\partial \varvec{\Pi }} \times \frac{\partial f}{\partial {\mathbf {P}}} \right) } \\&\quad - \varvec{\Gamma }\cdot {\left( \frac{\partial f}{\partial \varvec{\Pi }} \times \frac{\partial h}{\partial \varvec{\Gamma }} - \frac{\partial h}{\partial \varvec{\Pi }} \times \frac{\partial f}{\partial \varvec{\Gamma }} \right) }. \end{aligned}$$

The uncontrolled underwater vehicle from Example 3 is governed by the Lie–Poisson equation with respect to this bracket. Note also that the heavy top on a movable base after potential shaping shown in Example 7 is also described in terms of the same bracket.

1.3 Lie–Poisson bracket on \(({\mathfrak {s}} \ltimes (X \times Y))^{*}\)

Matching described in Sect. 4.3 yields Lie–Poisson equation on the extended \(({\mathfrak {s}} \ltimes (X \times Y))^{*}\) with the additional parameters living in \(Y^{*}\). Using the representation \(\tau \) defined in Sect. 4.3, we have the Lie–Poisson bracket on \(({\mathfrak {s}} \ltimes (X \times Y))^{*}\) as follows:

$$\begin{aligned} \left\{ f,h\right\} _{({\mathfrak {s}} \ltimes (X \times Y))^{*}}(\mu ,p,a,b)&= \left\{ f,h\right\} _{{\mathfrak {s}}^{*}} \!-{\left\langle a,\!\right\rangle }{ \sigma '{\left( \frac{\delta f}{\delta (\mu ,p)}\right) } \frac{\partial h}{\partial a} \!- \sigma '{\left( \frac{\delta h}{\delta (\mu ,p)}\right) } \frac{\partial f}{\partial a} } \nonumber \\&\quad - {\left\langle b, \tau '{\left( \frac{\delta f}{\delta (\mu ,p)}\right) } \frac{\partial h}{\partial b} - \tau '{\left( \frac{\delta h}{\delta (\mu ,p)}\right) } \frac{\partial f}{\partial b} \right\rangle }.\quad \end{aligned}$$
(34)

Example 12

(Lie–Poisson bracket on \((\mathfrak {se}(3) \ltimes ({\mathbb {R}}^{3} \times ({\mathbb {R}}^{4} \times {\mathbb {R}}^{4})))^{*}\)) Consider the case with \({\mathfrak {s}} = \mathfrak {se}(3)\), \(X = {\mathbb {R}}^{3}\), and \(Y = {\mathbb {R}}^{4} \times {\mathbb {R}}^{4}\). Using the expression for \(\tau '\) from (18), (34) gives, using the shorthand \(\Delta _{i} = (\varvec{\Delta }_{i},\delta _{i}) \in {\mathbb {R}}^{4}\) with \(i = 1, 2\),

$$\begin{aligned}&\left\{ f,h\right\} _{(\mathfrak {se}(3) \ltimes ({\mathbb {R}}^{3} \times ({\mathbb {R}}^{4} \times {\mathbb {R}}^{4})))^{*}}(\varvec{\Pi }, {\mathbf {P}}, \varvec{\Gamma }, \Delta _{1}, \Delta _{2}) \\&\quad =-\varvec{\Pi }\cdot {\left( \frac{\partial f}{\partial \varvec{\Pi }} \times \frac{\partial h}{\partial \varvec{\Pi }} \right) } - {\mathbf {P}}\cdot {\left( \frac{\partial f}{\partial \varvec{\Pi }} \times \frac{\partial h}{\partial {\mathbf {P}}} - \frac{\partial h}{\partial \varvec{\Pi }} \times \frac{\partial f}{\partial {\mathbf {P}}} \right) }\\&\qquad - \varvec{\Gamma }\cdot {\left( \frac{\partial f}{\partial \varvec{\Pi }} \times \frac{\partial h}{\partial \varvec{\Gamma }} - \frac{\partial h}{\partial \varvec{\Pi }} \times \frac{\partial f}{\partial \varvec{\Gamma }} \right) } \\&\qquad - \sum _{i=1}^{2} \varvec{\Delta }_{i} \cdot {\left( \frac{\partial f}{\partial \varvec{\Pi }} \times \frac{\partial h}{\partial \varvec{\Delta }_{i}} - \frac{\partial h}{\partial \varvec{\Pi }} \times \frac{\partial f}{\partial \varvec{\Delta }_{i}} - \frac{\partial f}{\partial \delta _{i}} \frac{\partial h}{\partial {\mathbf {P}}} + \frac{\partial h}{\partial \delta _{i}} \frac{\partial f}{\partial {\mathbf {P}}} \right) } \\&\quad = \frac{\partial f}{\partial \varvec{\Pi }} \cdot {\left( \varvec{\Pi }\times \frac{\partial h}{\partial \varvec{\Pi }} + {\mathbf {P}}\times \frac{\partial h}{\partial {\mathbf {P}}} + \varvec{\Gamma }\times \frac{\partial h}{\partial \varvec{\Gamma }} + \sum _{i=1}^{2} \varvec{\Delta }_{i} \times \frac{\partial h}{\partial \varvec{\Delta }_{i}} \right) } \\&\qquad + \frac{\partial f}{\partial {\mathbf {P}}} \cdot {\left( {\mathbf {P}}\times \frac{\partial h}{\partial \varvec{\Pi }} - \sum _{i=1}^{2} \frac{\partial h}{\partial \delta _{i}} \varvec{\Delta }_{i} \right) } + \frac{\partial f}{\partial \varvec{\Gamma }} \cdot {\left( \varvec{\Gamma }\times \frac{\partial h}{\partial \varvec{\Pi }} \right) } \\&\qquad + \sum _{i=1}^{2} {\left( \frac{\partial f}{\partial \varvec{\Delta }_{i}} \cdot {\left( \varvec{\Delta }_{i} \times \frac{\partial h}{\partial \varvec{\Pi }} \right) } + \frac{\partial f}{\partial \delta _{i}} {\left( \varvec{\Delta }_{i} \cdot \frac{\partial h}{\partial {\mathbf {P}}} \right) } \right) }. \end{aligned}$$

This is the Lie–Poisson bracket for the controlled system (29) from Example 9. One sees from the expression that \({\mathbf {P}}\cdot (\varvec{\Delta }_{1} \times \varvec{\Delta }_{2}), {\left\| \varvec{\Gamma }\right\| }^{2}\), \({\left\| \varvec{\Delta }_{i}\right\| }^{2}\), \(\varvec{\Gamma }\cdot \varvec{\Delta }_{i}\), \(\varvec{\Delta }_{1} \cdot \varvec{\Delta }_{2}\) with \(i = 1,2\) are Casimirs.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Contreras, C., Ohsawa, T. Controlled Lagrangians and stabilization of Euler–Poincaré mechanical systems with broken symmetry II: potential shaping. Math. Control Signals Syst. 34, 329–359 (2022). https://doi.org/10.1007/s00498-021-00312-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00498-021-00312-z

Keywords

Mathematics Subject Classification

Navigation