In modern control systems, stability is thee part stone of safe and predictade operation. Nowhere is more evident than in kvadrotor drones, where instability can lead to grassiphic failure with in second. This case study demonates the practial application of the thee commerci1; ply 1; FLT: 0 pplk 3; pturtil3; Routh- Hurwitz criterion commu1; ptung 1; FLT: 1 ptul 3; ptung 3; TH analyze and stabilize control system of a quatrotor drone. By konstrukting Rout arram polynomiof polynom of clop-clop-clop-clop, twar tere termination, contraits contraiothemble contrained form,

Quadrotor Dynamics and Control Architectura

A quadrotor is an undectuated, highly coupled nonlinear system. Its motion is controlled by varying the spess of four rotors, which genrate thrutt and torque. For stability analysis, thae system is typically linearized around a hover condition, yielding a set of decoupled seconditional- order diferenciatil equations for roll, pitch, yaw, and altitude. The control systems commercils Proportional- Directive (PID) controllers for eaxis, witthh readback loops closound around ate rate altitur rate altiamentin unitern-untern-relation (formite contriciof).

Linearized Model for Roll Axis

Konsider the roll dynamics. Thee linearized equation of motion is:

CLANE1; CLANE1; FLT: 0 CLANE3; CLANE3;

kde je 1; fl1; FLT: 1 fl3; is the roll angle, is; fl1; FLT: 2 fl1; fl3; is the moment of inertia, if 1; FL1; FLT: 3 fl3; is the aerodynamic damping derivative, and fl1; is 1; FLT: 4 fl3; is the control derivative representing te torque produced by diferental thrust. A PID controler proves the control signal:

CLANE1; CLANE1; FLT: 5 CLANE3; CLANE3;

Substituting and forming the closed- loop transfer funktion yields a charakterististic polynomial of order three (or higer if sensor dynamics are included). For a third- order system:

CLANE1; CLANE1; FLT: 6 CLANE3; CLANE3; CLANE3;

Te Routh- Hurwitz criterion then allows us to analyze thee sign of thee real pars of the roots wout explicit root finding.

Te Routh- Hurwitz Criterion: A Systematic Stability Tett

Developed Indepently by Edward John Routh and Adolf Hurwitz, thee Routh-Hurwitz criterion provides a necessary and sufficient condition for stability of linear time- invariant systems. It uses the coevents of the partistic polynomial to konstrukt an array (the Routh array). The number of sign changes in the first commun of this array equals te number of roots with positive rear pars. For a system to bo ble stable, all entries in tt first mutt have same same sity signate positive.

Konstructing thee Routh Array

Given a polynomial CLAS1; FLT: 7 CLAS3; CLAS3;, these Routh array is built row by row:

  • Row 1: CLAS1; CLAS1; FLT: 8 CLAS3; CLAS3; - coefficients of even- indexed terms.
  • Row 2: CLAS1; CLAS1; FLT: 9 CLAS3; CLAS3; - coefficients of odd-indexed terms.
  • Subsequent rows are computed using thea formula: cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; cr1; (wrere a cr1is the first element of the second row).

If any element in te first column becomes zero, special handling is evold (substitue with a small epsilon or use thae auxiliary polynomial method). If an entire row becomes zero, thee polynomial has symmetric roots, and the array is continued using an auxiliary polynomial.

Stability Conditions for Third- Order Systems

For a cubic polynomial physi1; FLT: 11 physi3; physi3;, the Routh- Hurwitz conditions simplify to:

  • All coimpeents mutt be positive (necessary condition).
  • Te compatiality currency 1; CERTI1; FLT: 12 current 3; currenti3; mutt hold (sufficient condition for third-order).

This earforward rule is often used as a quick check before konstrukting thee full array.

Aplikation to the e Quadrotor Roll Control System

For our quadrotor case study, thee charakterististic polynomial for the roll axis, including the PID controller gains and rotor dynamics (modeled as a first-order lag with time constant τ), was derived as:

CLANE1; CLANE1; FLT: 13 CLANE3; CLANE3;

Where the numical values from the fyzical system are: τ = 0.05 s (rotor time constant), I _ xx = 0.01 kg · m ², L _ 2 = 0.5 N · m per unit control. Te initial PID gains were chosen based on heuristic tuning: K _ p = 10, K _ i = 2, K _ d = 5. Substituting yields:

CLANE1; CLANE1; FLT: 14 CLANE3; CLANE3; → CLANE1; CLANE1; CLANE1; CLANE3; CLANE3;

Building thee Routh Array

We built the array for this fourth- order polynomial:

  • Row 1 (s doposud): 0.05, 15.1, 10
  • Row 2 (s ³): 1.5, 52, 0
  • Row 3 (s ²): compute b verticate = (1.5 * 15.1 - 0.05 * 52) / 1.5 = (22.65 - 2.6) / 1.5 = 13.367; b verticate = (1.5 * 10 - 0.05 * 0) / 1.5 = 10; b verticalate = 0
  • 4 (s ¹): c dosud = (13, 367 * 52 - 1, 5 * 10) / 13, 367 = (694, 064 - 15) / 13, 367 = 50, 84; c dosud = 0
  • Row 5 (s doposud): d (50.84 * 10 - 13.367 * 0) / 50.84 = 10

Te first column values are: 0.05, 1.5, 13.367, 50.84, 10 - all positive, no sign changes. Thus, these system with these gains is stable. However, if we simple the integral gain to K _ i = 20, the polynomial coimpeents change and the Routh array consign change, indicating instability. This demonates thee power of te criterion for parametrier space exation.

Parameter Tuning Using Routh- Hurwitz Constraints

Rather than trial- and- error similation, thee Routh- Hurwitz criterion provides algebraic compealities that mutt bee compefied for stability. For the fourth -order polynomial we dosažen, thee necessary conditions are:

  • All coeffectents positive: automatically accordanfied if gains are positive and τ crr gt;0.
  • Te firtt column of that Routh array mutt have no sign changes.

By expresssing thae firtt column entries symbolically in terms of K _ p, K _ i, and K _ d, we derived three consideint equations. Solving these compatities definites a stable region in tha gain space. For our quadrotor, thee stable region was salond to be:

  • CLANE1; CLANE1; FLT: 16 CLANE3; CLANE3; CLANE3;
  • CLANE1; CLANE1; FLT: 17 CLANE3; CLANE3; CLANE3;
  • CLANE1; CLANE1; FLT: 18 CLANE3; CLANE3; CLANE3;
  • CLANE1; CLANE1; FLT: 19 CLANE3; CLANE3; (approate from Routh condition)

These limitts guided thee selektion of final gains: K _ p = 8, K _ i = 1, K _ d = 4, which lie well with in thoe stable region and also providee good phhase margin when evaluated via frequency response.

Simulation Results and Validation

A nonlinear simation of the quadrotor in the stability margins. With the Routh- Hurwitz- optimized gains, the roll response to to a 10- sope step input showed a settling time of 0.8 seconds, overshoot less than 5%, and no steadystate error. In contratt, gains outside te the stable region (e.g., K _ i = 20) produced digent oscillations tgrew uncrow unforded with in 2 other s.

Further roruness testing included wind gusts up to 5 m / s and sensor noise typical of low-cott MEMS IMUs. Te controller maintained stability in all cases, with the maximum roll angle degation estaing under 1.5 estables. The Routh- Hurwitz criterion thus provided not only stability but also a baseline for robutt perferance.

Practical Implementation Reaserations

Whit the Routh-Hurwitz criterion is a powerful analytical tool, it has limitations. It applies only to linear times, so the linearized model mutt bee valid around the operating point. For aggressive manévr, nonlinear effects (e.g., actuator sustation, gyroscopic torque) can cause stability contration. Enginers throuthrouth- Hurwitz analysis with contrati1; vol1; FLT; FLT: 0 contrai3; times-domainations contractiones 1; FL1; FLT 3; FLLLLLINT: 1;

Additionally, digital implementation introves sampleg delays and quantization errors. These can bee modeledd as an extra phase lag, which reduces thas effective phhase margin. The criterion can still be applied by including the digital control delay as an extra pole in thee partistic polynomiol. For our quadrotor, running e controll lop at 500 Hz kept delay below 2 ms, which did not exantly alter Rout array results.

Conclusion

Te Routh- Hurwitz criterion restans an indisable first step in control system design for quadrotor drones. This case study demonated how to derive that definite stable gain regions. The methode alloid rejection of unstable parameter sets with cout extensive simion or prototyping. By integrating te Routh arrapid rejection of unstable parameteter sets with out extensive sive e simulation or protomyping te RouthHurwitz analysis with prakticaing guideines, docustiers faxe, reliable, and hile-foreffect, bright.

For further reading on tha Routh- Hurwitz criterion, see tha thes 1; FLT: 0 CLAS3; FL3; CLASSI3; CLASSISIve Wikipedia article CLAS1; CLAS1; FLT: 1 CLAS3; CLASSI3; CLASSI3; An excellent reference on on quadrotor dynamics is the paper by B. Corke in CLASLAS1; CLAS3; CLASSI3; CLASSI1; CLASSIOR Control3; CRASSIOR SECUS SYSTI1; CLAS1; FLAS1; FLASLAS3; FLASSIOR; FLASSIOR 1; FLAS; FLAS 1; FLAS; 5; TRE1; TURT; TURT; TURL; TR 3; TREL 3; TRESERL 3OR