AFIDLOS Path Tracking for Bathymetry USVs on Microcontrollers
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing path tracking methods for unmanned surface vehicles suffer from slow convergence, large overshooting, and poor anti-interference, particularly in complex water environments, due to the complexity of computational methods like observers and MPC controls, which are not suitable for microcontrollers.
Innovation Solution
An adaptive fuzzy integral differential line-of-sight (AFIDLOS) method is introduced, incorporating an integral term and differential term to counteract sideslip angles, a time-varying look-ahead distance, and a fuzzy controller to adjust convergence rates, combined with a linear quadratic regulator (LQR) controller for efficient path tracking on microcontrollers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional LOS guidance rate is used, then the control method is simple, but the convergence is slow and overshooting is large
Solution Approach 1:
The patent applies dynamics by making the look-ahead distance dynamic rather than fixed. The look-ahead distance is adjusted in real-time based on the lateral error and its derivative, allowing the system to adapt to different tracking stages. This dynamic adjustment enables faster convergence when error is large while preventing overshooting when error is small, resolving the contradiction between simple control and fast convergence.
Solution Approach 2:
The patent changes the parameter of look-ahead distance from a constant value to a variable that depends on lateral error and its rate of change. By modifying this key parameter dynamically, the system achieves improved convergence speed and reduced overshooting without requiring complex control algorithms, thus resolving the contradiction between simplicity and performance.
2Device complexity
If fixed look-ahead distance is used, then the control is simple, but the path tracking accuracy is poor
Solution Approach 1:
The patent transforms the fixed look-ahead distance parameter into a dynamic parameter that varies with lateral error and its derivative. This parameter change allows the system to maintain simple control structure while achieving high path tracking accuracy through adaptive adjustment of the look-ahead distance based on real-time tracking performance.
Solution Approach 2:
The patent introduces feedback by using lateral error and its derivative to adjust the look-ahead distance. This feedback mechanism enables the system to automatically adapt to tracking deviations, improving path tracking accuracy without complicating the overall control structure, thus resolving the contradiction between simplicity and precision.
3Manufacturing precision
If intelligent algorithms like particle swarm optimization or genetic algorithm are used to optimize controller parameters, then the path tracking accuracy is improved, but the computational power requirement increases making them unsuitable for microcontrollers
Solution Approach 1:
The patent replaces complex intelligent algorithms with a simpler, computationally efficient approach using basic calculus operations. Instead of requiring heavy computational resources for particle swarm optimization or genetic algorithms, the system uses analytical solutions based on derivatives, which are much lighter and suitable for microcontroller implementation while maintaining good path tracking accuracy.
Solution Approach 2:
The patent substitutes iterative intelligent optimization algorithms with an analytical mathematical approach. By using derivative-based look-ahead distance adjustment, the system replaces computationally intensive search methods with a direct calculation method, reducing computational complexity while achieving comparable or better performance suitable for embedded systems.
4Manufacturing precision
If model predictive control (MPC) is used, then the path tracking accuracy and convergence speed are improved, but the computational power requirement increases making it unfavorable for application in microcontrollers
Solution Approach 1:
The patent extracts the essential function of MPC (predicting future error trends) and implements it through a simplified derivative-based look-ahead distance adjustment. By taking out only the necessary predictive capability and removing the complex optimization and constraint handling of full MPC, the system achieves good path tracking accuracy with minimal computational requirements suitable for microcontrollers.
Solution Approach 2:
The patent replaces the computationally expensive MPC framework with a lightweight analytical solution. The derivative-based adjustment method provides similar predictive functionality without requiring iterative optimization, making it economically viable for implementation on resource-constrained microcontroller platforms while maintaining high path tracking accuracy.
Data Source
AI summary
An adaptive fuzzy integral differential line-of-sight (AFIDLOS) method for path tracking of a laser bathymetry unmanned surface vehicle is provided. The AFIDLOS method includes determining an AFIDLOS manner, establishing an unmanned surface vehicle control model, determining an LQR heading controller, and determining a path tracking manner by combining the AFIDLOS manner and the LQR controller to realize a path tracking control of an unmanned surface vehicle in a microcontroller. The method for path tracking is verified in experiments. Experimental results show that, compared with a traditional LOS guidance rate, 79.85% reduction in overshoot, and 55.32% shorter adjustment time are achieved by the AFIDLOS manner in simulation experiments, while 9.5% of an average lateral error is reduced in the Beihai Beach experiment, and an overlap rate between strips reaches 30% in the Pinqing Lake experiment, which meets the accuracy requirements of bathymetric mapping.