NORNAV-AUV – DEFI CLE : Robotique

NORNAV-AUV project

Nonlinear Robust Navigation Control of Small Autonomous Tethered UnderwaterVehicles

Control of unmanned underwater vehicles (UUVs) to accurately track a time-varying trajectory remains an open problem. This is due to their several inherent challenges, including (i) highly nonlinear and coupled dynamics (ii) time varying behavior, (iii) internal and external disturbances, (iv) parametric uncertainties, (v) unmodelled dynamics, and (vi) unpredictable operating environment. Based on these challenges, the present work will be focused on the tracking control problem of UUVs through the development of new nonlinear robust control schemes. Accordingly, the main objective will be to design and experimentally validate advanced robust control schemes for trajectory tracking of unmanned underwater vehicles in presence of unknown external disturbances and parametric uncertainties.

Target

1) A new version of a second order adaptive sliding mode control.

2) A new version of a disturbance observer based on the Extended State Estimator.

3) A new SMC scheme combined with a time delay estimator (TDE).

4) An adaptive version of the TDE-based SMC scheme.

Expected results

The goal of this collaboration was to allow the user (the pilot) to control an underwater robot in varying conditions, where the dynamic parameters of the system are changing very fast and where there are many strong sources of disturbances.

Type of financing

Incoming mobility

Researcherl

Jesus GUERRERO

Referent

Ahmed CHEMORI

Partners

Mexican Tecnológico Nacional de México

Keywords

Navigation control

Underwater vehicled

Autonomous

Varying conditions

Date

01/11/2022 au 01/02/2023

Tutelle

CNRS Montpellier

L’ équipe

Jesus GUERRERO

Researcher

Mexican Tecnológico Nacional de México
Team : ITS Abasolo
Automatic Control

Ahmed CHEMORI

Researcher

LIRMM
Team : DEXTER
Robotique médicale et mécanismes parallèles

+ Résumé du projet

The NORNAV-AUV project, focused on « locomotion and navigation, » aimed to develop robust control systems for autonomous underwater vehicles (UUVs) to enable a user to control an underwater robot in varying and disturbed conditions without requiring advanced piloting skills. This project is part of the « human-centered robotic challenge » and meets the expectations of « field robotics » by developing controllers for piloting assistance in underwater environmental assessment.
Underwater vehicles present significant control challenges due to their complex and nonlinear dynamics, difficulties in estimating hydrodynamic parameters, random external disturbances, and unmodeled dynamics. Currently, most commercial underwater vehicles or experimental prototypes use Proportional-Derivative (PD) or Proportional-Integral-Derivative (PID) controllers due to their simplicity of implementation and tuning. However, these controllers show limitations in the face of parametric uncertainties, time-varying external disturbances, and require gain adjustments under different operating conditions.

To address these challenges, four new robust control strategies were designed and applied to the « Leonard » underwater vehicle at LIRMM:
1. Adaptable Second-Order Sliding Mode Control: Based on sliding mode theory, this scheme offers advantages such as finite-time convergence, insensitivity to time-varying external disturbances, and reduction of the « chattering » phenomenon in the control input. An algorithm was introduced for the autonomous self-tuning of gains, providing an advantage when controlling UUVs in real environments without requiring fine-tuning.

2. New Finite-Time Disturbance Estimator: A disturbance observer based on the Extended State Observer (ESO) and Sliding Mode Control (SMC) was proposed. This approach combines the structural simplicity of the ESO with the robust finite-time estimation capability of the SMC. The idea is to integrate this proposal with non-robust controllers such as PDs, still widely used in Remotely Operated Vehicles (ROVs).

3. SMC Control with Time-Delayed Estimator (TDE): The Time-Delayed Estimation (TDE) technique is used to estimate hydrodynamic parameters in real-time using information provided by onboard sensors and a mathematical model of the underwater vehicle. The TDE information is introduced into the second-order SMC as equivalent control, offering the advantage of reducing « chattering » in the control input.

4. Saturated SMC Controls: Second-order SMC controls are implemented taking into account actuator saturation issues. Control laws are modified to prevent input saturation, allowing the vehicle to operate safely in real environments.

All proposed control solutions were tested in real-time on the « Leonard » underwater vehicle at LIRMM. The test pool within the robotics laboratory was used to recreate realistic scenarios for four experimental tests: a nominal scenario without external disturbances, a robustness test against parametric uncertainties, a robustness test against sudden mass variation, and a robustness test against external disturbances. The results were refined for the preparation of scientific articles in high-level specialized journals, with four articles submitted and three undergoing a second round of review.

+ Référence du projet

1. Saturated STA-based sliding mode tracking control of AUVs. Submitted to Ocean Engineering Journal (Q1). Impact Factor : 5.0. https://hal-lirmm.ccsd.cnrs.fr/lirmm-04516120v1

2. Robust Super Twisting Algorithm-based disturbance observer for autonomous underwater vehicles: design, stability analysis, and real-time experiments. Submitted to Robotics and Autonomous Systems (Q1). Impact Factor : 4.3. https://hal-lirmm.ccsd.cnrs.fr/lirmm-04782129v1

3. STA-based design of an adaptive disturbance observer for autonomous underwater vehicles: from concept to real-time validation. Submitted to Control Engineering Practice (Q1). Impact Factor : 4.9. https://hal-lirmm.ccsd.cnrs.fr/lirmm-04364453v1

4. Improved Adaptive High-Order Sliding Mode Based Control for Trajectory Tracking of Autonomous Underwater Vehicles. IEEE Journal of Oceanic Engineering (Q1). Impact Factor : 4.7. https://hal-lirmm.ccsd.cnrs.fr/lirmm-04605914v1

+ Références – état de l’art

¹ D. Forchelet et al., “Enclosed electronic system for force measurements in knee implants,” Sensors (Switzerland), vol. 14, no. 8, pp. 15009–15021, 2014.

² F. X. Wang et al., “Novel Force Measurement System for Soft Tissue Balance in Total Knee Arthroplasty Based on Flexible Pressure Sensor Arrays,” Adv. Intell. Syst., p. 2100156, 2022.

³X. Sun et al., “Development of a Flexible Meniscus Sensor for Detection of Force Distribution in Knee Arthroplasty,” IEEE Sens. J., vol. 20, no. 2, pp. 700–706, 2020.

⁴ H. Jiang, et al., “A wireless visualized sensing system with prosthesis pose reconstruction for total knee arthroplasty,” Sensors (Switzerland), vol. 19, no. 13, 2019.

⁵ H. Sun et al., “A soft thumb-sized vision-based sensor with accurate all-round force perception,” Nature Machine Intelligence, 4(2), 135–145, 2022.

⁶ Y. C. Huang et al., “Sensitive pressure sensors based on conductive microstructured air-gap gates and two-dimensional semiconductor transistors,” Nat. Electron., vol. 3, no. 1, pp. 59–69, 2020.

⁷ O. A. Araromi et al., “Ultra-sensitive and resilient compliant strain gauges for soft machines,” Nature, vol. 587, no. 7833, pp. 219–224, Nov. 2020.

⁸ C. M. Boutry et al., “A hierarchically patterned, bioinspired e-skin able to detect the direction of applied pressure for robotics,” Sci. Robot., vol. 3, no. 24, p. eaau6914, 2018.

⁹ X. Zeng et al., “A bioinspired three-dimensional integrated e-skin for multiple mechanical stimuli recognition,” Nano Energy, vol. 92, p. 106777, Feb. 2021.

¹⁰W. Yuan, S. Dong, and E. H. Adelson, “GelSight: High-resolution robot tactile sensors for estimating geometry and force,” Sensors (Switzerland), vol. 17, no. 12, 2017.

¹¹ D. Ma, E. Donlon, S. Dong, and A. Rodriguez, “Dense Tactile Force Distribution Estimation using GelSlim and inverse FEM,” ICRA 2019, 2018.

¹² T. Hellebrekers, N. Chang, K. Chin, M. J. Ford, O. Kroemer, and C. Majidi, “Soft Magnetic Tactile Skin for Continuous Force and Location Estimation Using Neural Networks,” IEEE Robot. Autom. Lett., vol. 5, no. 3, pp. 3892–3898, Jul. 2020.

¹³T. P. Tomo et al., “Covering a Robot Fingertip With uSkin: A Soft Electronic Skin With Distributed 3-Axis Force Sensitive Elements for Robot Hands,” IEEE Robot. Autom. Lett., vol. 3, no. 1, pp. 124–131, 2018.

¹⁴ H. Wang et al., “Design methodology for magnetic field-based soft tri-axis tactile sensors,” Sensors (Switzerland), vol. 16, no. 9, 2016.

¹⁵ Y. Yan et al., “Soft magnetic skin for super-resolution tactile sensing with force self-decoupling,” Sci. Robot., vol. 6, no. 51, p. eabc8801, Feb. 2021.