Sponsor: U.S. National Science Foundation

Project No.: CMMI-2226936

Duration: Jan 1, 2023 – Dec 31, 2025

Principal Investigator: Professor Shaoping Xiao

Past and Current Graduate Students: Junchao Li, and Soheyla Tofighi

Summary

This grant is supporting research that contributes new knowledge related to control and dynamical systems, promoting the application of artificial intelligence in engineering problem-solving. Extreme weather events like heavy rains are expected to occur frequently with continued climate change. The resulting disasters (e.g., flooding) increase the damage and impacts on economics, national security, public health, and human well-being. This award supports research in developing a computer program that can intelligently operate reservoirs to reduce flood risk as much as possible. Various artificial intelligence techniques are applied so that the computer program can learn the best operation strategy, considering incomplete data acquisition, weather uncertainty, and ethical decision-making. In addition, this project includes undergraduate research opportunities and some outreach activities for K-12 students. In particular, this project will develop educational materials for K-6 students to learn how climate change affects people’s lives.

Related Publications, Presentations, and Other Outcomes:

Xiao, S., LI, J., and Wang, Z., "Model-free motion planning of complex tasks subject to ethical constraints," 26^th International Conference on Human-Computer Interaction, Washington DC, USA, June 29 – July 4, 2024

Li, J., Cai, M., and Xiao, S. P., "Reinforcement learning-based motion planning in partially observable environments under ethical constraints," AI and Ethics, 2024, https://doi.org/10.1007/s43681-024-00441-6

Li, J., Cai, M., Kan, Z., and Xiao, S. P., "Model-free reinforcement learning for motion planning of autonomous agents with complex tasks in partially observable environments," Autonomous Agents and Multi-agent Systems, 38(14), 2024, https://doi.org/10.1007/s10458-024-09641-0

Gurbuz, F., Mudireddy, A., Mantilla, R., and Xiao, S., "Using a physics-based hydrological model and storm transposition to investigate machine-learning algorithms for streamflow prediction," Journal of Hydrology, 628, 2023, 130504, https://doi.org/10.1016/j.jhydrol.2023.130504

Li, J. C., Cai, M., Wang, Z. A., and  Xiao, S. P., “Model-based motion planning in POMDPs with temporal logic specifications”, Advanced Robotics, 37(14), 2023, 871-886, https://doi.org/10.1080/01691864.2023.2226191

S. Xiao, and J. Li, “Model-free control synthesis of autonomous agents in partially observable and dynamic environments”, 2^nd IACM Mechanistic Machine Learning and Digital Engineering for Computational Science Engineering and Technology, El Paso, TX, September 24-27, 2023

J. Li, and S. Xiao, “Robotics Motion Planning for Complex Tasks in Partially Observable Environments Using Model-free Reinforcement Learning”, 2023 IMECE- International Mechanical Engineering Congress & Exposition, New Orleans, LA, October 29 – November 2, 2023

Model-based reinforcement learning algorithms and codes: https://github.com/JunchaoLi001/LDGBA-Model_Checking

Model-based reinforcement learning algorithms and codes: https://github.com/JunchaoLi001/Model-free_DRL_LSTM_on_POMDP_with_LDGBA

Reinforcement learning with ethical constraints algorithms and codes: https://github.com/JunchaoLi001/Model-free_DRL_POMDP_LDGBA_with_ethical_constraints

AI Youth camp: https://xiao.lab.uiowa.edu/ai-youth-camp

Sponsor: U.S. National Science Foundation

Project No.: CMMI-2104383

Duration: Aug 15, 2021 – Feb 28, 2025

Principal Investigator: Professor Shaoping Xiao

Co Principal Investigators: Professors Caterina Lamuta and Phillip Deierling

Past and Current Graduate Students: Siamak Attarian, Arunabha Batabyal, Yingbin Chen, Akram Ghaffarigharehbagh, and Mahmudul Alam Shakib

Summary: This project is developing a new computer model to design and study the mechanical behavior of metal-ceramic composites. In most traditional composites, the component distributions are constant to enhance the material properties evenly. In contrast, the metal-ceramics composites studied in this project have component distributions that vary from location to location. Therefore, engineers can design the proper composite structure with various desired material properties at different critical locations. The new computer model includes several computation algorithms at nanoscale, microscale, and macroscale, respectively. In addition, the project provides for validation of the computer model by fabricating the composite samples and testing them on different types of equipment. This project also consists of some outreach activities for undergraduate students and K-12 students.

Related Publications, Presentations, and Other Outcomes:

Attarian, S., Xiao, S. P., “Development of a 2NN-MEAM potential for Ti-B system and studies of the temperature dependence of the nanohardness of TiB2”, Computational Materials Science, 201, 2022, 110875. https://doi.org/10.1016/j.commatsci.2021.110875

El Tuhami, A. and Xiao S. P. “Multiscale Modeling of Metal-Ceramic Spatially Tailored Materials via Gaussian Process Regression and Peridynamics”, International Journal of Computational Methods, 2022, 2250025. https://doi.org/10.1142/S0219876222500256

Attarian, S. and Xiao, S. P., “Investigating the strength of Ti/TiB interfaces at multiple scales using density functional theory, molecular dynamics, and cohesive zone modeling”, Ceramic International, 48(22), 2022, 33185-33199. https://doi.org/10.1016/j.ceramint.2022.07.259

S. Xiao, “Multiscale modeling of metal-ceramic spatially tailored materials via machine learning,” Engineering Mechanics Institute Conference 2022, Baltimore, Maryland, May 31-June 3, 2022

Xiao, S., Attarian, S., and Deierling, P. “Deeping learning in multiscale modeling of spatially tailored materials,” 15^th World Congress on Computational Mechanics, Yokohama, Japan, July 31-Aug 5, 2022

Xiao, S. “Investigating the mechanics of Ti/TiB interfaces at multiple scales: from quantum mechanics to molecular dynamics”, 4^th International Conference on Materials Science and Engineering, Houston, TX, April, 2023

Xiao, S. Li, J., Bordas, S. P. A., and Kim, T. Y., “Artificial neural networks and their applications in computational materials science: A review and a case study,” Advanced in Applied Mechanics, 57, 2023, 1-33, https://doi.org/10.1016/bs.aams.2023.09.001

Dataset and neural networks: https://github.com/jwli0728/ANNs-in-Material-Science

AI Youth Summer Camp: https://xiao.lab.uiowa.edu/ai-youth-camp