• No results found

Pareto Seeking Control: On-line multi-objective optimization of dynami systems


Academic year: 2021

Share "Pareto Seeking Control: On-line multi-objective optimization of dynami systems"

Show more ( Page)

Full text


On-line Multi-Objective Optimization of

Dynamic Systems: Pareto seeking control

Khalid Tourkey Atta∗ Andreas Johansson∗ Thomas Gustafsson∗

Department of Computer Science, Electrical and Space Engineering,

Lule˚a University of Technology, Lule˚a, Sweden .

Abstract: Considering a Multi-Objective Optimization problem for a Dynamic System, we propose a controller that drives this system into a point on the Pareto front.At such a point, no change in any control variable can be made without deteriorating one of the objectives. Gradient Estimators are the core of this controller. A sinusoidal Perturbation based gradient estimation approach is adopted.

Keywords: Pareto Seeking Control, Multi Objective Optimization, Pareto Front


The multi-objective optimization problem can be defined as the task of maximizing (or minimizing) a number of functions (also called objectives) subject to a number of constraints (or limitations) by adjusting a number of design (decision) variables. This idea presents a concept more than being a definition due to the absence of a unique single solution for this problem(Marler and Arora (2004)). Instead of the optimal solution (point), the term Pareto optimal is used in the multi-objective optimization problem. In simple words a Pareto optimal point is a point such that there is no other point that does not deteriorate at least one objective and at the same time it represent an optimal solution for one of the objectives. By Relaxing the condition of optimality of one of the objectives (Marler and Arora (2004)), the set of all weakly Pareto optimal points is obtained which is also termed the Pareto Frontier (or Front or Surface) and represents a trade off between the different objectives (Lampinen (2000); Caramia and Dell’Olmo (2008)).


Consider a non linear system with input u(t) ∈ Rm and

a number of objectives to be optimized y(t) ∈ Rl. The

problem of optimizing the objectives can be stated in a manner similar to the Extremum seeking control Problem (Zhang and Ord´o˜nez (2007)) as:

maximize u y(t) = [h1(x(t)), h2(x(t)), . . . hl(x(t))] subject to gi(x(t), u(t)) ≤ 0, j = 1, . . . p, dxi(t) dt − fi(x(t), u(t)) = 0, i = 1, . . . n, x(0) = x0 (1)

where x = [x1, x2, . . . , xn]T ∈ Rn is a vector representing

the state variables with x0 representing the initial

con-dition of the states variable. Also, f = [f1, f2, . . . , fn]T,

h = [h1, h2, . . . , hl]T, and g = [g1, g2, . . . , gp]T are

vector-valued functions. Both f , h and g are assumed to be

sufficiently smooth.

Our goal is to find a controller that drives the system into a point within the Pareto front. This controller will be termed the Pareto Seeking Controller (PSC).


The idea of the proposed PSC is to adjust u(t) in a direction that improves all the objectives, and stops when reaching a point such that any small change in u will cause a reduction in one of the objectives. The algorithm requires knowledge of the gradient of all the objectives at the current operation point. Gradient estimators are discussed in the Extremum seeking problem, and one of the proposed estimators can be used in our scenario (Gelbert et al. (2012); Krsti´c and Wang (2000))

For the case of two objectives, the system is at the Pareto front if the gradient of the two objectives are in opposite direction. Then, the suitable direction is found by adding the normalized gradients:

du0 i dt = k ∂y¯ 1 ∂ui + ¯ ∂y2 ∂ui  , i = 1, 2 (2) In this way, the direction will be zero for all the points on the Pareto front, and no further changes are made. Fig. 1 shows the case of a plant with two inputs and two objectives. Plant k1/s a1 sin(ω1t) y1 Gradient Estimator-Normalized Output y2 u1 o u1 k2/s a2 sin(ω2t) u2 o u2 Gradient Estimator-Normalized Output 1 1 u y   2 1 u y   1 2 u y   2 2 u y  


It should be noted that, the point that the system will converge is not unique, and will depend mainly on the initial condition.


A plant with two inputs and two objective outputs were considered. The plant is represented as a cascade combi-nation of linear dynamics and a static nonlinearity (Krsti´c (2000)), similar to Fig 2, with the following dynamics:



u θ Ψ(θ) y

Fig. 2. An approximated non Linear System

Fi(s) =    1 0.33s + 1 0 0 1 0.5s + 1   , Fo(s) =    1 s + 1 0 0 1 0.5s + 1   , Ψ(θ) = ψ1(θ) ψ2(θ)  ,

A perturbation based gradient estimator were used to find the gradient. Two cases were simulated, the first case is a maximization problem with the non-linear functions as:

ψ1(θ) = (θ − θ1∗) T3 0 0 3  (θ − θ1∗) ψ2(θ) = (θ − θ2∗) T1 0 0 1  (θ − θ2∗)

where θ1∗ = [−2, +2]T and θ∗2 = [+2, −2]T. Different

starting points were simulated. The left part of Fig. 3 shows how the controlled variables are converging to the Pareto front, while the right part shows the objective functions increasing until the Pareto front is reached. The second case is a minimization problem with non linear function (Lampinen (2000)):

ψ1(θ) = θ(1)2+ θ(2), ψ2(θ) = θ(1) + θ(2)2

Fig. 3. First case, Left: u1 and u2and the convergence to

the Pareto Frontier of the input variables, Right: y1

and y2 and the convergence to the Pareto Frontier of

the Objectives

Fig. 4. Second case, Left: u1 and u2 and the convergence

to the Pareto Frontier of the input variables, Right: y1and y2and the convergence to the Pareto Frontier

of the Objectives

Fig. 4 shows the convergence to the Pareto Front for different cases.


This proposed controller is shown to drive the system to the Pareto front for two examples. Similar to the Extremum Seeking Control, a detailed knowledge of the system dynamics is not necessary. Only an estimate of the gradient of the output is required.


Caramia, M. and Dell’Olmo, P. (2008). Multi-objective Management in Freight Logistics. Springer.

Gelbert, G., Moeck, J., Paschereit, C., and King, R. (2012). Advanced algorithms for gradient estimation in one-and two-parameter extremum seeking controllers. Journal of Process Control.

Krsti´c, M. (2000). Performance improvement and limita-tions in extremum seeking control. Systems & Control Letters, 39(5), 313–326.

Krsti´c, M. and Wang, H. (2000). Stability of extremum seeking feedback for general nonlinear dynamic systems. Automatica, 36(4), 595–601.

Lampinen, J. (2000). Multiobjective nonlinear pareto-optimization. Pre-investigation Report, Lappeenranta University of Technology.

Marler, R.T. and Arora, J.S. (2004). Survey of multi-objective optimization methods for engineering. Struc-tural and multidisciplinary optimization, 26(6), 369–395. Zhang, C. and Ord´o˜nez, R. (2007). Numerical optimization-based extremum seeking control with ap-plication to abs design. Automatic Control, IEEE Trans-actions on, 52(3), 454–467.


Related documents

Riksdagen ställer sig bakom det som anförs i motionen om att se över möjligheten att installera solpaneler på taken till samtliga av riksdagens byggnader och tillkännager detta

Vägmarkeringsklass Nuvarande krav Föreslaget krav 1 På varje kilometersträcka bör högst 30% av mätplatserna ha ett retroreflexionsmedelvärde under- stigande 100 mcd/m 2 /lux

To describe from the patient self-reported recovery profile, the prevalence of postoperative item variables after cytoreductive surgery and hyperthermic intraperitoneal

Carolines ambition från början var att göra ett bra material för barn med läs och skrivsvårigheter och att det nu fungerar att arbeta inom många områden som hjälp är ett

Då det var sökningarna på gruppen Beröring och gruppen Demens som resulterade i artikelresultatet och utfallet på antal artiklar inte påverkades av sökordet ”Quality of

Musiken i grundskolan har en uppgift att ge eleverna en inkörsport till ämnet musik, vilket framförallt är viktigt för de elever som inte får intresset med sig

Measured unfolded differential cross sections of W W production in the eµ final state for the transverse momentum of the leading lepton, p lead. T , the invariant mass, m `` , and

Institute of High Energy Physics, Chinese Academy of Sciences, Beijing; (b) Department of Modern Physics, University of Science and Technology of China, Anhui; (c) Department


In addition to the driving behavior parameters, lane changing distance and parameters for conflict areas were adjusted during the calibration. The default lane change distance for a

Enligt kalkylbladet Nytta under resa beror den slutliga nyttovinsten under resan på följande faktorer: andelen som byter rutt vid stor störning, genomsnittlig tidsvinst för dessa

Publication fractions (P; full counts within parentheses), MNCS, PP(top 10%), MNJS and MJPPtop10% by department and for the whole publication period 2010-2017... Faculty of

Det skulle vara användbart ur flera olika perspektiv att undersöka och förstå mer exakt vilka transdiagnostiska faktorer som förekommer vid stress- och smärtproblematik samt

To understand whether OPT deliberations give room for instrumental interpretations, this study assessed whether citizen proposals were used transparently; firstly in what stage of

When a fault is injected into a relay or relay position sensor, the model can not discriminate between the correct candidate and the wires leading to it from the I/O hardware.. This

To best of our knowledge, using a contemporary method, this is the first study to quantify the location and extent of pain in a population with a high proportion of older people aged

Edin och Vesterlund menar att det fanns spår av paternalism inom Public service begreppet och att detta var ett hinder för att mer demokratiska relationer skulle kunna

Vilket går att relatera till Skolverkets rapport från 2005, där det framgår att vårdnadshavare väljer att ha sina barn kortare tid i fritidshemmet då de anser

Figure 4.11: Comparison CDFs for deriving best value, change number of particle(s) reset position For RPSO, as it is shown in Figure 4.11 we selected different number of particle

So far, most studies have used Das and Dennis’s structured approach for generating a uniformly distributed set of reference points on the unit simplex.. Due to the highly

The results presented in this thesis demonstrate that human dermal FBs possess stem cell plasticity, and can be induced to differentiate into cells with phenotypes

This thesis work aims at investigating the trade-off offered by optimal control techniques between energy consumption and reference tracking for torque allocation to the

Corresponding to these types of energy resources are eight types of energy supply systems: biofuel CHP (BCHP), wind turbine (WT), solar photovoltaic (PV), solar PV/thermal collector