⚡ electrical engineering

Modeling and Simulation of an Active Car Suspension with a Robust LQR Controller under Road Disturbance, Parameter Uncertainty and White Noise

This study demonstrates that an active car suspension system controlled by a robust Linear Quadratic Regulator (LQR) outperforms both passive and PID-controlled systems in terms of ride comfort, safety, and stability under road disturbances, parameter uncertainties, and white noise, as evidenced by its superior rise time, overshoot, and settling time characteristics.

Original authors: Mehmet Karahan

Published 2026-02-06

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Mehmet Karahan

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine your car's suspension system as a pair of legs trying to walk on a bumpy, uneven path. The goal is to keep the upper body (the passengers and the car's cabin) as smooth and still as possible, even when the feet (the wheels) are hitting rocks and potholes.

This paper is essentially a race between three different "leg" strategies to see which one keeps the rider most comfortable and safe.

The Three Competitors

The Passive Suspension (The Old School Jogger):
Think of this as a person with stiff, unchangeable knees and ankles. They have springs and shock absorbers, but they can't react to the terrain. If you hit a bump, they just bounce up and down until the energy naturally fades away. It's cheap and simple, but if the road gets rough, the ride gets bumpy.
The PID-Controlled Active Suspension (The Smart Jogger with a Coach):
This is a person with a coach shouting instructions. The coach (a PID controller) watches the bumps and tells the legs to push or pull to counteract them. It's better than the stiff knees, but the coach is using a fairly standard, reactive method. It's like trying to balance a broomstick on your hand using only your eyes and a basic rule of thumb.
The LQR-Controlled Active Suspension (The Olympic Athlete with a Supercomputer):
This is the star of the show. This suspension system also has an actuator (a motor) to push and pull, but it's controlled by a "Linear Quadratic Regulator" (LQR). Imagine this as an Olympic athlete connected to a supercomputer that calculates the perfect amount of force needed at every single millisecond. It doesn't just react; it anticipates and optimizes the movement to minimize effort and maximize smoothness.

The Obstacle Course (The Experiments)

The researchers didn't just test these on a smooth track. They put all three through a rigorous obstacle course in a computer simulation (using MATLAB/Simulink) to see how they handled different challenges:

The Sudden Pothole: A single, sharp bump in the road.
The Heavy Load: They added extra weight to the car (like filling it with passengers and luggage) to simulate "parameter uncertainty." This is like asking the joggers to run while carrying a heavy backpack they weren't expecting.
The Chaotic Road: They added "white noise," which is like driving on a road that is constantly vibrating with tiny, random jitters, not just big bumps.

The Results: Who Won the Race?

The researchers measured three things to judge the winners:

Rise Time: How fast did the car react to the bump?
Overshoot: Did the car bounce too high or too low before settling? (Like a trampoline that bounces you too high).
Settling Time: How long did it take for the car to stop shaking and return to a smooth ride?

The Verdict:

The LQR-controlled active suspension won every single category.

Speed: It reacted the fastest to the bumps.
Control: It had the least "overshoot." While the other systems bounced up and down a bit too much before calming down, the LQR system barely wobbled.
Stability: It settled down to a smooth ride much quicker than the others. Even when the car was heavy or the road was full of random noise, the LQR system remained the most stable.

The Bottom Line

The paper concludes that while the old "Passive" system is simple and the "PID" system is a decent upgrade, the LQR-controlled system is the clear champion.

Think of it this way: If the Passive system is a wooden cart and the PID system is a bicycle with good suspension, the LQR system is a high-tech hovercraft that glides over the bumps. The study proves that by using this advanced mathematical controller, you get a ride that is not only safer (because the tires stay glued to the road) but also significantly more comfortable for the passengers, regardless of how bumpy the road gets or how heavy the car is.

The researchers also checked the "mathematical heartbeat" (pole-zero maps) of all three systems and confirmed that they are all stable (they won't fall apart), but the LQR system is simply the most efficient and robust at its job.

Technical Summary: Modeling and Simulation of an Active Car Suspension with a Robust LQR Controller

Problem Statement
Vehicle suspension systems are critical for ensuring passenger comfort, road holding, and safety by mitigating vibrations and shocks from road surfaces. While passive suspension systems are widely used due to their simplicity and low cost, they lack actuators and controllers, limiting their ability to adapt to varying road conditions. Active suspension systems, which incorporate actuators and controllers, offer superior performance but require complex design and higher energy consumption. Although Proportional-Integral-Derivative (PID) controllers are commonly employed in active suspensions for their ease of adjustment, there is a need to evaluate more robust control strategies that can maintain performance under diverse and challenging conditions, including road disturbances, parameter uncertainties (such as variable vehicle load), and environmental noise.

Methodology
This study modeled a quarter-car active suspension system using MATLAB/Simulink to compare three configurations: a classical passive suspension, an active suspension controlled by a PID controller, and an active suspension controlled by a Linear Quadratic Regulator (LQR).

System Modeling: The dynamic equations of motion for the quarter-car model were derived, defining state variables including suspension travel, sprung mass velocity, wheel deflection, and wheel vertical velocity.
Controller Design:
- PID Controller: Two PID controllers were implemented (one for sprung mass motion and one for suspension travel). Coefficients ( $K_p, K_i, K_d$ ) were tuned using the Simulink PID Tuner feature.
- LQR Controller: An optimal control approach was designed to minimize a quadratic cost function. The weighting matrices $Q$ (state cost) and $R$ (control effort) were determined using Bryson's rule to penalize errors in state variables and control effort, respectively. The gain matrix $K$ was calculated by solving the Riccati equation.
Simulation Scenarios: The performance of all three systems was evaluated under four distinct conditions:
- Stability Analysis: Pole-zero maps were generated to verify system stability.
- Road Disturbance: A step disturbance (0.08 m amplitude) was applied at a vehicle speed of 72 km/h.
- Parameter Uncertainty: Simulations were repeated with a +20% increase in sprung mass to simulate varying passenger/cargo loads.
- White Noise: Band-limited white Gaussian noise was added to the road disturbance to simulate continuous, random road irregularities.
Performance Metrics: The systems were compared based on rise time, overshoot, and settling time for three key outputs: suspension travel, sprung mass acceleration, and sprung mass motion.

Key Contributions
The paper distinguishes itself from existing literature through three primary contributions:

Comprehensive Stability Analysis: Unlike many studies that rely solely on graphical performance comparisons, this work generates pole-zero maps for all three suspension types (passive, PID-active, and LQR-active) to conduct a detailed stability analysis, confirming that all systems remain stable with poles located to the left of the imaginary axis.
Quantitative Robustness Comparison: The study moves beyond graphical analysis by providing specific numerical time-response data (rise time, overshoot, settling time) to compare the robustness of the controllers under different operating conditions.
Multi-Condition Evaluation: While many studies focus only on instantaneous road disturbances, this research evaluates suspension performance under simultaneous road disturbance and parameter uncertainty, as well as under road disturbance with white noise, offering a more rigorous test of controller robustness.

Results
The simulation results consistently demonstrated the superiority of the LQR-controlled active suspension over both the passive and PID-controlled active systems across all tested scenarios:

Under Road Disturbance: The LQR controller achieved the fastest rise time (e.g., 1.15 s for suspension travel vs. 1.29 s for passive), the least overshoot, and the shortest settling time (1.21 s vs. 1.451 s for passive).
Under Parameter Uncertainty (+20% Mass): The LQR system maintained its advantage, showing the fastest rise time and shortest settling time for all metrics, indicating robustness against load variations.
Under White Noise: Even with added noise, the LQR controller provided the fastest rise time and shortest settling time. For instance, in sprung mass motion, the LQR system settled in 2.872 s compared to 2.901 s for the passive system.
General Trend: In every category (suspension travel, sprung mass acceleration, and sprung mass motion) and under every condition, the LQR-controlled active suspension exhibited the least oscillation and the quickest return to equilibrium, whereas the passive system consistently showed the longest settling times and highest overshoots.

Significance and Claims
The paper concludes that the LQR-controlled active suspension system provides a more comfortable and safe ride compared to both passive suspension and PID-controlled active suspension. The study claims that the LQR approach is the most successful configuration, evidenced by its ability to minimize overshoot and achieve the fastest settling times under varying road disturbances, parameter uncertainties, and noise. The authors assert that these simulation results prove the robustness and superior performance of the LQR controller in enhancing vehicle dynamics and passenger comfort.

The Three Competitors

The Obstacle Course (The Experiments)

The Results: Who Won the Race?

The Bottom Line

More like this