Research Funding

Collaborative Research: Elements: Discrete Simulation of Flexible Structures and Soft Robots

PI: M. Khalid Jawed (UCLA), co-PI: Jungseock Joo (UCLA), Andrew Sabelhaus (Boston U.), Carmel Majidi (Carnegie Mellon University)

Funding Agency: Software Institutes, National Science Foundation

Dates: 10/01/22 - 09/30/25

Award Number: 2209782, 2209784, 2209783

LINK: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2209782&HistoricalAwards=false

Abstract: The objective of this work is to develop a discrete differential geometry (DDG) simulation environment into a widely-available software package capable of modeling soft and flexible structures. The DDG approach enables low-dimensional modeling of slender rods and flexible shells combined into arbitrary shapes, establishing a practical but still physically accurate contrast to computationally expensive finite element analysis (FEA) techniques. This work first develops a core software package for DiSMech that adapts prior work to meet the standard for national cyberinfrastructure: maintainable, extensible, and with a robust user interface. Next, a virtual testbed for a wide class of soft and flexible robots is built by incorporating DiSMech into an existing robotics software suite. The project team will use the combined software framework with a machine learning approach to develop a locomotion strategy for example soft robots. Finally, add-ons to DiSMech will incorporate machine learning alongside the DDG-based physics models for even faster simulations, demonstrating the research potential for this software in uncovering underlying physical phenomena. By advancing DDG-based physics simulations to capture a wide range of soft and flexible structures, with a computational speed sufficient for learned robot control, all in an easy-to-use interface, DiSMech addresses an important gap in the national cyberinfrastructure.

CCRI: Planning-C: A Framework for Development of Robots and IoT for Precision Agriculture

PI: M. Khalid Jawed (UCLA), co-PI: Sriram Narasimhan (UCLA), Jungseock Joo (UCLA), Wei Wang (UCLA)

Funding Agency: CISE Community Research Infrastructure, National Science Foundation

Dates: 08/15/22 - 01/31/24

Award Number: 2213839

LINK: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2213839&HistoricalAwards=false

Abstract: The fourth industrial revolution - characterized by smart automation and inter-connectivity - is about to change farm management practices forever. To hasten this positive change, the project team envisions an infrastructure that enables rapid development of robotic hardware, sensing technologies, software tools, and machine learning algorithms. While computer scientists and roboticists are developing novel software and hardware every day with great potential for precision agriculture, these tools typically fall short of real-world application. The envisioned simulation environment for testing such tools will bridge the gap between fundamental research and real-world deployment. These tools can enable autonomous farm management, including precision weed/pest management, precision irrigation, autonomous crop health monitoring, and precision crop protection. This can dramatically cut down labor costs, reduce chemical usage, lower the impact on water resources, conserve the fertility of soil, and increase the yield of crops. Computer scientists and roboticists will be able to get familiar with real-world challenges of precision agriculture, e.g., dramatic effects of precipitation on agriculture. The project team will collect preliminary data to prototype the infrastructure, organize workshops with interested researchers to plan the infrastructure, and connect with farmers and agronomists to gather feedback.

DSFAS: Harnessing Data for Accurate Yield and Oil Content Prediction

PI: Wei Wang (UCLA), co-PI: M. Khalid Jawed (UCLA), Joao Paulo Flores (NDSU), Mukhlesur

Rahman (NDSU)

Funding Agency: National Institute of Food and Agriculture, United States Department of Agriculture

Dates: 06/15/22 - 06/14/26

Award Number: 2022-67022-37021

LINK: https://portal.nifa.usda.gov/web/crisprojectpages/1028284-dsfas-harnessing-data-for-accurate-yield-and-oil-content-prediction.html 

Abstract: The research objective is design, development, and field-testing of an artificially intelligent method to predict the yield and oil content of flax from a number of morphological traits before harvesting. Success in this project will result in a quantum leap for flax breeding programs by bringing in a systematic data-driven autonomous approach, in lieu of conventional heuristic-based decision-making. NDSU hosts the only flax breeding program in the US and North Dakota is the largest producer of flax (91% of US production), which will be used as the testbed in this project. The project requires precision data collection on morphological traits throughout the entire life cycle of the crop. A cooperative team of unmanned aerial systems (UASs) and unmanned ground vehicles (UGVs) will be employed. The UGVs will also be loaded with a hyperspectral camera to predict the oil content even before harvesting. The scheduling and operation of the UAS-UGV team will be dictated by a data analytics engine. The videos collected by the UAS/UGV will be processed to extract various morphological traits and to predict the final yield of the crop through an integrated machine learning model. Hyperspectral images will be analyzed using machine learning to predict the oil content of each plot. If successful, this predictive model for yield and oil content will allow a breeder to substantially lower the costs of the breeding program, and hereby improve the quality of the new crop variety. In collaboration with AmeriFlax, this framework will be tested in real-world settings.

Machine Learning-based Modeling and Operation for Smart Machining of Aluminum Aerospace Structures

PI: Xiaochun Li (UCLA), co-PI: M. Khalid Jawed (UCLA), T. C. Tsao (UCLA)

Funding Agency: Clean Energy Smart Manufacturing Innovation Institute (funded by U.S. Department of Energy)

Dates: 02/01/22 - 01/31/23

LINK: https://www.cesmii.org/ 

Abstract: In this project, we aim at making two MAE machine operations, the drilling machine and milling machine smarter. The data and model systems focus on modeled systems with phenomena not easily modeled with first principles models, e.g. machining processes that affect part quality, for example, drilling and milling as well as the situational thermal and mechanical impacts on the precision of a part and energy and productivity efficiency of the processes.

Collaborative Research: Mechanics of Knots and Tangles of Elastic Rods

PI: M. Khalid Jawed (UCLA), co-PI: Bashir Khoda (University of Maine)

Funding Agency: Mechanics of Materials and Structures, National Science Foundation

Dates: 10/01/21 - 09/30/24 

Award Number: 2101751, 2101745

LINK: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2101751&HistoricalAwards=false

Abstract: The research objective of this project is to quantify the mechanical response of knots tied in elastic rods. The project will employ (1) fast numerical simulations inspired by computer graphics, (2) innovative materials with customizable friction, and (3) autonomous robotic experiments to untangle the mechanics of knots. Even in the case of the most basic type of knots (overhand knots), the force required to tie the knot depends on an intricate interplay of (1) elasticity, (2) friction, and (3) topology. Interestingly, the overhand knot may undergo a snap-through buckling instability beyond a critical amount of pull. Such instability in a basic knot points to the richness of the mechanical behavior of knots. After developing simulation and experimental tools, the mechanical response and instabilities of a few common knots, e.g. overhand and shoelace knots, will be investigated. Exploiting the computational speed of the simulation tool and autonomy of robotic experiments, the mechanical response of several types of knots will be quantified to build a library of their mechanics. This data will be used to rationalize the variation of a knot’s mechanical response as a function of the topological, material, and frictional parameters. Similar to the periodic table of elements, a mechanics-based classification scheme of knots will be formulated, where the knots will be grouped into various classes, such as, friction-dominated knots, bending-dominated knots, and others.

CAREER: MaLPhySiCS - Machine Learning-assisted Physics-based Simulation and Control of Soft Robots

PI: M. Khalid Jawed (UCLA)

Funding Agency: Dynamics, Control and System Diagnostics (DCSD), National Science Foundation

Dates: 10/01/21 - 09/30/26

Award Number: 2047663

LINK: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2047663&HistoricalAwards=false

Abstract: The research objective of this project is formulation of fast and efficient physics-based simulation, assisted by machine learning, for autonomous control of soft robots. Using this framework, a macroscale bacteria-inspired robot will be designed and controlled. This robot will use buckling in flagellum (thin flexible tail) to control its swimming direction. This is expected to be the simplest autonomous soft robot with a single scalar control input. Two key challenges to be tackled in the project are: (1) computational efficiency so that the simulation can be used for optimization, and (2) physical accuracy and robustness of the models so that model-based control can be employed on the real robotic systems. Towards this goal, machine learning-assisted modeling of complex systems in a discrete differential geometry-based simulation framework is planned. Neural network-based models for the structure of the robot and the hydrodynamics will be developed. These models are expected to be as fast as simplified heuristic models and as accurate as physics-based fine-grained models. This simulation tool will be used to develop a model-based control framework for the bacteria-inspired robot for untethered autonomous operation. This robot can help us gain insight into bacterial locomotion, e.g., role of instability in bacterial propulsion. From a robotics perspective, the robot has only one control input with minuscule number of moving parts. The design of the robot makes it amenable for miniaturization to sub-millimeter scale with potential biomedical applications.

Deep Spring: a Neural Network-based Approach to Design of Slender Structures

PI: Vwani Roychowdhury (UCLA), co-PI: M. Khalid Jawed (UCLA)

Funding Agency: Computational & Data-Enabled Science & Engineering, National Science Foundation

Dates: 07/01/21 - 06/30/24

Award Number: 2053971

LINK: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2053971&HistoricalAwards=false

Abstract: This project will enable the design of new materials and structures like soft kirigami composites and compressive buckling-induced micro-sized 3D architecture materials. The researchers will advance numerical modeling of systems with slender structures through a combination of machine learning and discrete differential geometry, using “nonlinear springs” represented by neural networks to simulate the systems. The enhanced computational speed of this approach will aid in the design and optimization of engineering systems such as deployable structures and soft robots. Developing predictive models for slender structures and metamaterials is challenging because of the inherent instabilities and many possible configurations. If successful, this development would be the first of its kind in the mechanics research community and would provide the twin capability to predict complex structural responses and design structures with an inverse problem solver. While the target proof-of-concept examples relate to slender structures and metamaterials, the computational methods would be generalizable to a broad class of material-structure systems.

Autonomous Robotic Systems for Precision Weed Control in Flax

PI: Mukhlesur Rahman (NDSU), co-PI: Mohiuddin Quadir (NDSU) and M. Khalid Jawed (UCLA)

Funding Agency: National Institute of Food and Agriculture, United States Department of Agriculture

Dates: 02/01/21 - 01/31/25

Award Number: 2021-67022-34200

Abstract: North Dakota is the leading producer of flax for oil and food use. The crop is a poor competitor with in-field weeds that reduce seed production. Traditionally, herbicides are used for weed control using large-scale sprayers that dispense herbicide over crops and weeds alike. The method results in the overuse of herbicides and a reduction in crop yield. In the United States, 1.2 billion pounds of pesticides were used in agricultural crops, and the cost was about $14 billion in 2012. Agricultural Robotics can be used for the precision application of pesticides that may reduce 90% of total pesticide use and reduce material costs by $12.6 billion. In addition to the money saved, optimized pesticide use can reduce environmental contamination and increase the organic life of the soil. In this project, we are attempting to develop a small, low-cost autonomous weed-spraying rover capable of computer vision-based weed detection, complete field coverage, and automatic recharge in an experimental field of flax crops. The proposed robot has four major systems, (i) a robot control and stability system, (ii) a computer vision-based navigation system, (iii) a weed identification system, and (iv) an herbicide application system. In addition, robot-actuable spray solutions will be formulated, which will lower the amount of herbicide sprayed per unit area of the field and show targeted efficiency in suppressing weed growth. The target application is hypothesized to make weed control in flax and other row crop fields fully autonomous, even when the spacing between crop lines is as small as one foot.

NRI: FND: Physics-based Training of Robots for Manipulation of Ropes and Clothes

PI: M. Khalid Jawed (UCLA), co-PI: Jungseock Joo (UCLA)

Funding Agency: National Robotics Initiative, National Science Foundation

Dates: 09/01/19 - 08/31/23

Award Number: 1925360

LINK: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1925360&HistoricalAwards=false

Abstract: The research objective of this project is to fundamentally understand robotic manipulation of flexible objects (ropes & clothes) using model-based training. The team of researchers will develop physics-based simulation tools for the mechanics of deformable structures and demonstrate the application of fast and efficient simulations to train robots. To overcome the barriers associated with translating models to the real world, the researchers will use simulations, in conjunction with optimization, to formulate policies that are robust against uncertainties, e.g., friction and material defects. The goal is simulation-based training of cobots that is ready for application in the real world; this will largely remove the painstaking training process by physical demonstration required for collaborative robots. The strength of this approach will be demonstrated through autonomous folding of towels and tying of knots to secure objects.

Multimodal Communicative Learning for Robot Navigation

PI: Jungseock Joo (UCLA), co-PI: M. Khalid Jawed (UCLA)

Funding Agency: Amazon (Science Hub for Humanity and Artificial Intelligence)

Dates: 2022

LINK: https://www.sciencehub.ucla.edu/

Abstract: We propose to develop a comprehensive learning framework for autonomous navigating robots that can communicate and interact with human users while navigating indoors and outdoors. Robotic navigation has received much attention in industry and academia and has many real world applications such as food delivery robots and household robots (e.g., Amazon Astros). Our proposed research specifically tackles three key challenges in robotic navigation as follows.

• Multimodal Human Interaction

• Scalable and Customizable Learning

• Simulation-to-Real