A Novel Data-Driven MPC Framework Using KDE and KPCA for Autonomous Vehicles
Modeling and controlling complex, nonlinear, large-scale systems such as autonomous vehicles presents significant challenges due to high dimensionality, uncertain dynamics, and real-time constraints. This paper introduces a novel data-driven predictive control framework that synergistically integrat...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11018417/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849418043005861888 |
|---|---|
| author | Romdhane Nasri Jannet Jamii Majdi Mansouri Zouhaier Affi Vicenc Puig |
| author_facet | Romdhane Nasri Jannet Jamii Majdi Mansouri Zouhaier Affi Vicenc Puig |
| author_sort | Romdhane Nasri |
| collection | DOAJ |
| description | Modeling and controlling complex, nonlinear, large-scale systems such as autonomous vehicles presents significant challenges due to high dimensionality, uncertain dynamics, and real-time constraints. This paper introduces a novel data-driven predictive control framework that synergistically integrates Kernel Density Estimation (KDE), Kernel Principal Component Analysis (KPCA), and Hankel matrix representation to address these challenges. The Hankel matrix formulation captures temporal correlations in system dynamics, enabling an effective representation of state transitions and dynamic modes. This representation is complemented by KPCA for nonlinear dimensionality reduction and KDE for probabilistic uncertainty quantification, resulting in a comprehensive framework that maintains computational tractability without sacrificing model fidelity. We present a rigorous mathematical foundation for the integration of these techniques, establishing formal convergence guarantees and error bounds for the kernel-based approximations. The framework demonstrates remarkable efficiency in feature extraction, preserving 95% of system variance with only two principal components, while the KDE component provides robust probabilistic predictions essential for safe autonomous navigation. Extensive experimental validation on an autonomous vehicle platform yields outstanding performance metrics: lateral tracking errors of 0.0225 m (RMSE), heading errors of 0.0476 rad, and longitudinal velocity tracking errors of 0.2725 m/s. These results represent an 81.3–89.8% improvement over traditional MPC approaches across diverse operational scenarios, including urban navigation, highway driving, parking maneuvers, and obstacle avoidance. The computational efficiency of our approach (1.6 ms execution time) enables real-time implementation on standard automotive-grade embedded control units while maintaining 100% compliance with lateral safety constraints. Comprehensive comparative analysis against state-of-the-art methods demonstrates that our framework effectively addresses the fundamental challenges of nonlinearity, high dimensionality, uncertainty quantification, and real-time implementation in autonomous vehicle control. These contributions advance the field of data-driven control for complex, nonlinear, large-scale systems, offering a principled alternative to traditional model-based approaches with significant practical implications for autonomous driving technologies. |
| format | Article |
| id | doaj-art-166eea9a25da49e68f33e2b374f795d2 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-166eea9a25da49e68f33e2b374f795d22025-08-20T03:32:33ZengIEEEIEEE Access2169-35362025-01-011310963810965610.1109/ACCESS.2025.357528311018417A Novel Data-Driven MPC Framework Using KDE and KPCA for Autonomous VehiclesRomdhane Nasri0Jannet Jamii1Majdi Mansouri2https://orcid.org/0000-0001-6390-4304Zouhaier Affi3https://orcid.org/0009-0003-1303-2634Vicenc Puig4https://orcid.org/0000-0002-6364-6429Laboratory of Mechanical Engineering, National School of Engineers of Monastir, Monastir, TunisiaDepartment of Electrical Engineering, College of Engineering, King Faisal University, Al Hofuf, Saudi ArabiaDepartment of Electrical and Computer Engineering, College of Engineering, Sultan Qaboos University, Muscat, OmanLaboratory of Mechanical Engineering, National School of Engineers of Monastir, Monastir, TunisiaAdvanced Control Systems Research Group, Institut de Robòtica, CSIC-UPC, Universitat Politècnica de Catalunya-BarcelonaTech, Barcelona, SpainModeling and controlling complex, nonlinear, large-scale systems such as autonomous vehicles presents significant challenges due to high dimensionality, uncertain dynamics, and real-time constraints. This paper introduces a novel data-driven predictive control framework that synergistically integrates Kernel Density Estimation (KDE), Kernel Principal Component Analysis (KPCA), and Hankel matrix representation to address these challenges. The Hankel matrix formulation captures temporal correlations in system dynamics, enabling an effective representation of state transitions and dynamic modes. This representation is complemented by KPCA for nonlinear dimensionality reduction and KDE for probabilistic uncertainty quantification, resulting in a comprehensive framework that maintains computational tractability without sacrificing model fidelity. We present a rigorous mathematical foundation for the integration of these techniques, establishing formal convergence guarantees and error bounds for the kernel-based approximations. The framework demonstrates remarkable efficiency in feature extraction, preserving 95% of system variance with only two principal components, while the KDE component provides robust probabilistic predictions essential for safe autonomous navigation. Extensive experimental validation on an autonomous vehicle platform yields outstanding performance metrics: lateral tracking errors of 0.0225 m (RMSE), heading errors of 0.0476 rad, and longitudinal velocity tracking errors of 0.2725 m/s. These results represent an 81.3–89.8% improvement over traditional MPC approaches across diverse operational scenarios, including urban navigation, highway driving, parking maneuvers, and obstacle avoidance. The computational efficiency of our approach (1.6 ms execution time) enables real-time implementation on standard automotive-grade embedded control units while maintaining 100% compliance with lateral safety constraints. Comprehensive comparative analysis against state-of-the-art methods demonstrates that our framework effectively addresses the fundamental challenges of nonlinearity, high dimensionality, uncertainty quantification, and real-time implementation in autonomous vehicle control. These contributions advance the field of data-driven control for complex, nonlinear, large-scale systems, offering a principled alternative to traditional model-based approaches with significant practical implications for autonomous driving technologies.https://ieeexplore.ieee.org/document/11018417/Kernel-based predictive controlnonlinear and large-scale systemskernel density estimationkernel PCAHankel matrixautonomous vehicles |
| spellingShingle | Romdhane Nasri Jannet Jamii Majdi Mansouri Zouhaier Affi Vicenc Puig A Novel Data-Driven MPC Framework Using KDE and KPCA for Autonomous Vehicles IEEE Access Kernel-based predictive control nonlinear and large-scale systems kernel density estimation kernel PCA Hankel matrix autonomous vehicles |
| title | A Novel Data-Driven MPC Framework Using KDE and KPCA for Autonomous Vehicles |
| title_full | A Novel Data-Driven MPC Framework Using KDE and KPCA for Autonomous Vehicles |
| title_fullStr | A Novel Data-Driven MPC Framework Using KDE and KPCA for Autonomous Vehicles |
| title_full_unstemmed | A Novel Data-Driven MPC Framework Using KDE and KPCA for Autonomous Vehicles |
| title_short | A Novel Data-Driven MPC Framework Using KDE and KPCA for Autonomous Vehicles |
| title_sort | novel data driven mpc framework using kde and kpca for autonomous vehicles |
| topic | Kernel-based predictive control nonlinear and large-scale systems kernel density estimation kernel PCA Hankel matrix autonomous vehicles |
| url | https://ieeexplore.ieee.org/document/11018417/ |
| work_keys_str_mv | AT romdhanenasri anoveldatadrivenmpcframeworkusingkdeandkpcaforautonomousvehicles AT jannetjamii anoveldatadrivenmpcframeworkusingkdeandkpcaforautonomousvehicles AT majdimansouri anoveldatadrivenmpcframeworkusingkdeandkpcaforautonomousvehicles AT zouhaieraffi anoveldatadrivenmpcframeworkusingkdeandkpcaforautonomousvehicles AT vicencpuig anoveldatadrivenmpcframeworkusingkdeandkpcaforautonomousvehicles AT romdhanenasri noveldatadrivenmpcframeworkusingkdeandkpcaforautonomousvehicles AT jannetjamii noveldatadrivenmpcframeworkusingkdeandkpcaforautonomousvehicles AT majdimansouri noveldatadrivenmpcframeworkusingkdeandkpcaforautonomousvehicles AT zouhaieraffi noveldatadrivenmpcframeworkusingkdeandkpcaforautonomousvehicles AT vicencpuig noveldatadrivenmpcframeworkusingkdeandkpcaforautonomousvehicles |