Filtering by
- All Subjects: engineering
- All Subjects: Robots
- Creators: Marvi, Hamidreza
All civilization requires some sort of infrastructure to provide an essential service. Roads, bridges, pipelines, railroads, etc. are all critical in maintaining our society, but when they fail, they pose a serious threat to the economy, public safety, and environment. This is why it has become increasingly important to invest in and research the field of Structural Health Monitoring (SHM) to ensure the safety and reliability of our infrastructure. This research paper delves into the optimization of a Lizard-inspired Tube Inspection (LTI) robot, with the primary focus on the inspection side of SHM through the use of Electro Magnetic Acoustic Transducer (EMAT), a Non-Destructive Testing (NDT) method. The robot is designed to inspect power plants piping for damage or defects, and its ability to detect issues early, results in improved plant efficiency, enhanced structural data collection, and increased safety. An iterative, reliable design was constructed by reducing the weight and addressing previous design flaws and then tested. Solidworks was used to calculate theoretical weight, applied stress, and displacements for the design modifications.. The overall reduction in weight was around 12.4% of the previous design. While this research successfully reduced the robot's weight and resolved issues in its design, further optimization is still necessary. Future studies should investigate the finger and friction pad design, robot control, and ways to reduce the reliance on commercial off-the-shelf parts. This will expand the robot’s inspection capabilities, making it applicable in other industries where NDT is critical to ensure structural integrity and safety, such as the pipes in oil and gas refineries, water treatment plants, and chemical processing plants, innovating the way infrastructure is monitored and maintained.
In this paper, we discuss the methods and requirements to simulate a soft bodied beam using traditional rigid body kinematics to produce motion inspired by eels. Eels produce a form of undulatory locomotion called anguilliform locomotion that propagates waves throughout the entire body. The system that we are analyzing is a flexible 3D printed beam being actively driven by a servo motor. Using the simulation, we also analyze different parameters for these spines to maximize the linear speed of the system.
The robustness of a neural network is defined as the stability of the network output under small input perturbations. It has been shown that neural networks are very sensitive to input perturbations, and the prediction from convolutional neural networks can be totally different for input images that are visually indistinguishable to human eyes. Based on such property, hackers can reversely engineer the input to trick machine learning systems in targeted ways. These adversarial attacks have shown to be surprisingly effective, which has raised serious concerns over safety-critical applications like autonomous driving. In the meantime, many established defense mechanisms have shown to be vulnerable under more advanced attacks proposed later, and how to improve the robustness of neural networks is still an open question.
The generalizability of neural networks refers to the ability of networks to perform well on unseen data rather than just the data that they were trained on. Neural networks often fail to carry out reliable generalizations when the testing data is of different distribution compared with the training one, which will make autonomous driving systems risky under new environment. The generalizability of neural networks can also be limited whenever there is a scarcity of training data, while it can be expensive to acquire large datasets either experimentally or numerically for engineering applications, such as material and chemical design.
In this dissertation, we are thus motivated to improve the robustness and generalizability of neural networks. Firstly, unlike traditional bottom-up classifiers, we use a pre-trained generative model to perform top-down reasoning and infer the label information. The proposed generative classifier has shown to be promising in handling input distribution shifts. Secondly, we focus on improving the network robustness and propose an extension to adversarial training by considering the transformation invariance. Proposed method improves the robustness over state-of-the-art methods by 2.5% on MNIST and 3.7% on CIFAR-10. Thirdly, we focus on designing networks that generalize well at predicting physics response. Our physics prior knowledge is used to guide the designing of the network architecture, which enables efficient learning and inference. Proposed network is able to generalize well even when it is trained with a single image pair.
To achieve this goal, a model of a swarm performing a collective transport task in a bounded domain featuring convex obstacles was simulated in MATLAB/ Simulink®. The closed-loop dynamic equations of this model were linearized about an equilibrium state with angular acceleration and linear acceleration set to zero. The simulation was run over 30 times to confirm system ability to successfully transport the payload to a goal point without colliding with obstacles and determine ideal operating conditions by testing various orientations of objects in the bounded domain. An additional purely MATLAB simulation was run to identify local minima of the Hessian of the navigation-like potential function. By calculating this Hessian periodically throughout the system’s progress and determining the signs of its eigenvalues, a system could check whether it is trapped in a local minimum, and potentially dislodge itself through implementation of a stochastic term in the robot controllers. The eigenvalues of the Hessian calculated in this research suggested the model local minima were degenerate, indicating an error in the mathematical model for this system, which likely incurred during linearization of this highly nonlinear system.