UB AlertEmergency Information (7-Jun-23 6:15 pm) Air quality health advisory extended through Thursday, June 8.

Vision-Based Planning and Control in Support of Autonomous Inspection, Maintenance and Repair

Investigators

Xiao Liang, PhD
PhD, University of California, Berkeley
242 Ketter Hall
Phone: (716) 645-2298
liangx@buffalo.edu
Assistant Professor
Department of Civil, Structural and Environmental Engineering
School of Engineering and Applied Sciences
Call Xiao Liang Email Xiao Liang Profile for Xiao Liang
242 Ketter Hall
Phone: (716) 645-2298
liangx@buffalo.edu

Students: PhD Candidate Seyedomid Sajedi

Project description

Seyed Omid Sajedi, a PhD candidate in the Department of Civil, Structural and Environmental Engineering, kicked off the presentations with his research on artificially intelligent systems for rapid post-earthquake inspections.

Inspecting and maintaining operation infrastructure is crucial for its safety and sustainability. In order to provide maintenance and repairs to infrastructure, agencies generally rely on construction crews equipped with machinery, often under limited protection. Typically, these repairs require a partial or full shutdown, and result in costly disruptions to the public. Robots can be leveraged to perform maintenance and repairs because of their mixed mobility and reconfigurability in a challenging work environment. Xiao Liang identifies four tasks to integrate computer vision models with robotic systems to automate inspection and maintenance.

The first task uses unmanned aerial vehicles (UAV) to target the structural components for inspection. Task two uses the feedback from UAVs to develop planning and control algorithms to enable a flight path so the target structural component can be reached. The third tasks takes the input from the UAV and identifies and delineates visual changes along with evaluation sensors to determine if the structure is repairable. The final task is various maintenance and repair actions.

This project is funded by the Institute of Bridge Engineering (IBE), Region II University Transportation Center.

Chen, Z., Liang, X., & Zheng, M. (2021). Iterative Learning for Heterogeneous Systems. IEEE/ASME Transactions on Mechatronics, early access.
Sajedi, S. O., & Liang, X. (2021). Uncertainty-Assisted Deep Vision Structural Health Monitoring. Computer-Aided Civil & Infrastructure Engineering, 36(9), 1168-1184.
Zheng, M., Lyu, X., Liang, X., & Zhang, F. (2021). A Generalized Design Method for Learning-Based Disturbance Observer. IEEE/ASME Transactions on Mechatronics, 26(1), 45-54.
Chen, Z., Liang, X., & Zheng, M. (2020). Knowledge Transfer between Different UAVs for Trajectory Tracking. IEEE Robotics and Automation Letters, 5(3), 4939-4946.
Liang, X. (2019). Image-based post-disaster inspection of reinforced concrete bridge systems using deep learning with Bayesian optimization. Computer-Aided Civil & Infrastructure Engineering, 34(5), 415-430.
Liang, X., M. Zheng, M., & Zhang, F. (2018). A scalable model-based learning algorithm with application to UAVs. IEEE Control Systems Letters, 2(4), 839-844. (also presented at 57th Conference on Decision & Control)

Chen, Z., Liang, X., & Zheng, M. (2021, May). Deep Iterative Learning Control for Quadrotor's Trajectory Tracking. In 2021 American Control Conference (ACC) (pp. 1408-1413). IEEE.
Liang, X., & Sajedi, S. O. (2020). Deep Bayesian U-Nets for Efficient, Robust and Reliable Post-Disaster Damage Localization. 17th World Conference on Earthquake Engineering (17WCEE).
Sajedi, S. O., & Liang, X. (2019). A convolutional cost-sensitive crack localization algorithm for automated and reliable RC bridge inspection. 10th New York City Bridge Conference, New York City, NY.
Zheng, M., Chen, Z., & Liang, X. (2019). A preliminary study on a physical model oriented learning algorithm with application to UAVs. Dynamics System and Control Conference, Park City, UT.