Model- and Data-Based Control of Self-Balancing Robots: Practical Educational Approach with LabVIEW and Arduino

Overview

• This paper compares the performance of model-based and data-based control strategies for two-wheeled self-balancing robots (TWSBRs).
• Model-based control (MBC) algorithms, like PID control, require detailed dynamic modeling and mathematical manipulation.
• Data-based control (DBC) methods, like fuzzy control, provide a simpler approach without needing in-depth understanding of the system model.
• The paper illustrates the advantages and disadvantages of both MBC and DBC using a TWSBR.
• All controllers were tested on a cost-effective OSOYOO self-balancing kit.

Plain English Explanation

Two-wheeled self-balancing robots are tricky to control because they are non-linear and unstable systems. This study looks at two different ways to control these robots: model-based control (MBC) and data-based control (DBC).

MBC algorithms, like PID control, require a deep understanding of the robot's dynamics and complex math to get the control equations right. On the other hand, DBC methods, like fuzzy control, use machine learning to figure out how to control the robot without needing to know all the details of how it works.

The researchers tested both MBC and DBC controllers on a small, affordable self-balancing robot kit. They wanted to see how well each approach could keep the robot balanced and moving in the right direction. The results highlight the trade-offs between the two control strategies and show that both can be implemented on a budget-friendly platform.

Technical Explanation

The paper evaluates the performance of model-based control (MBC) and data-based control (DBC) approaches for controlling two-wheeled self-balancing robots (TWSBRs). MBC algorithms, such as Lead-Lag and PID control, require detailed dynamic modeling and complex mathematical manipulation to derive the appropriate control laws for the linearized equations of motion.

In contrast, DBC methods like fuzzy control provide a simpler and more intuitive way to design effective controllers without needing an in-depth understanding of the system model. The researchers implemented and tested various MBC and DBC controllers on the OSOYOO self-balancing robot kit, which includes an Arduino microcontroller, MPU-6050 sensor, and DC motors.

The control laws and user interface were developed using the LabVIEW-LINX toolkit, and a real-time hardware-in-the-loop experiment was conducted to validate the results. The study highlights the trade-offs between the two control approaches and demonstrates that both MBC and DBC strategies can be effectively implemented on a cost-effective platform.

Critical Analysis

The paper provides a practical, educational comparison of model-based and data-based control approaches for two-wheeled self-balancing robots. By testing the controllers on a low-cost hardware platform, the researchers have made the research more accessible and applicable to a wider audience, including educators and hobbyists.

However, the paper does not delve deeply into the specific implementation details or performance metrics of the individual control strategies. It would be helpful to see more quantitative data, such as settling times, overshoot, and energy efficiency, to better understand the relative strengths and weaknesses of the MBC and DBC approaches.

Additionally, the paper does not address potential issues with sensor noise, disturbances, or other real-world factors that could impact the performance of the controllers in practice. Further research may be needed to explore the robustness and reliability of these control strategies in more challenging environments.

Conclusion

This study offers a valuable comparison of model-based and data-based control strategies for two-wheeled self-balancing robots, a common platform in robotics education and research. The results demonstrate that both MBC and DBC approaches can be effectively implemented on a cost-effective hardware setup, providing educators and hobbyists with practical insights into the trade-offs between these control methods.

The research highlights the importance of considering the balance between the complexity of the control system and the depth of understanding required, which can be particularly relevant when designing educational robotics projects or building low-cost robotic platforms. By exploring these control strategies, the paper contributes to the ongoing efforts to make robotics more accessible and engaging for a wider range of learners and enthusiasts.