Researchers have developed a novel adaptive asymptotic tracking control system for vehicle active suspensions, integrating an event-triggered mechanism and actuator saturation compensation. The technology improves ride smoothness and control accuracy while cutting computational load and communication resource consumption by over 80%, providing a practical low-cost solution for next-generation intelligent vehicle suspensions. The findings are published in Advanced Equipment .Researchers have developed a novel adaptive asymptotic tracking control system for vehicle active suspensions, integrating an event-triggered mechanism and actuator saturation compensation. The technology improves ride smoothness and control accuracy while cutting computational load and communication resource consumption by over 80%, providing a practical low-cost solution for next-generation intelligent vehicle suspensions. The findings are published in Advanced Equipment .
Vehicle suspension systems are fundamental to ride comfort and driving safety: they absorb shocks from uneven road surfaces and keep the vehicle body stable. Compared with traditional passive suspensions, active suspensions can adjust their damping force in real time based on road conditions, delivering far better ride quality. However, conventional active suspension control relies on continuous signal sampling and calculation, which wastes on-board computing and communication resources. In addition, suspension actuators have inherent physical limits; extreme road conditions can easily push them beyond capacity, causing performance drops or even system instability.
To address these challenges, a team led by Associate Professor Yingjie Deng and Mr. Fangcheng Liu from Yanshan University, in collaboration with scholars from Mokpo National Maritime University in the Republic of Korea, has developed an innovative event-triggered adaptive neural control scheme for active suspension systems, achieving a balance between high control precision and low resource consumption.
“This design makes active suspensions work in a smarter way,” says Mr. Fangcheng Liu, corresponding author of the study. “Instead of updating control signals nonstop, the system only adjusts its output when body vibration reaches a preset threshold. This eliminates a huge amount of unnecessary computation and data transmission, while still maintaining a smooth and stable ride.”
One core innovation of the scheme is its built-in handling of actuator saturation—the maximum force the suspension actuator can physically output. The team designed a dedicated auxiliary system to compensate for saturation effects, preventing actuator overload even when driving over severe bumps. This greatly improves the reliability and practicality of the controller for real vehicles, where actuator capacity is always constrained.
The team also applied a minimal learning parameter technique paired with neural networks. Unlike traditional adaptive control methods that require updating dozens of parameters during operation, this design only needs to update one single parameter, drastically reducing computational complexity. The embedded neural network automatically adapts to unknown road conditions, working reliably from smooth highways to rough rural roads.
The performance of the technology was verified through both simulation and hardware experiments. The team tested the system under three typical road scenarios: bump roads, random uneven roads and step roads. Results showed that compared with existing fuzzy observer control schemes, the new method reduced steady-state tracking error by nearly 60%, and cut total control energy consumption by more than 90%. Thanks to the event-triggered mechanism, control signal updates were reduced by over 80% compared with traditional continuous sampling, with no loss in control performance. The team also confirmed that all key safety constraints—including suspension travel limits and tire grounding safety—are fully satisfied.
In bench tests on a hydraulic suspension experimental platform, the control scheme maintained excellent tracking accuracy under both square-wave and sine-wave vibration inputs, outperforming mainstream existing control methods in both stability and precision.
“This technology does more than just improve ride comfort,” notes Professor Yingjie Deng. “By lowering computing and communication loads, it reduces the hardware requirements for vehicle controllers, which helps cut costs in mass production. It can be applied to passenger cars, new energy vehicles and heavy commercial vehicles.”
Beyond automotive suspensions, the control framework can also be extended to other mechanical systems requiring vibration control with limited computing or communication resources, such as industrial precision platforms and rail vehicle suspensions.
The team notes that the current study is based on a quarter-car suspension model. For future work, they plan to expand the scheme to full-vehicle suspension systems, and add optimization for signal transmission delays to better adapt to complex real-vehicle network environments.
The paper “Adaptive asymptotic tracking control for active suspension systems: an event-triggered approach with input saturation” was published in Advanced Equipment .
Deng Y, Liu F, Meng X, Wu Y, Zhao D, et al. Adaptive asymptotic tracking control for active suspension systems: an event-triggered approach with input saturation. Adv. Equip . 2026(1):0005, https://doi.org/10.55092/ae20260005.
Advanced Equipment
Experimental study
Not applicable
Adaptive asymptotic tracking control for active suspension systems: an event-triggered approach with input saturation
8-Jul-2026