Department of Mechanical Engineering
Permanent URI for this community
The Virginia Tech Mechanical Engineering Department serves its students, alumni, the Commonwealth of Virginia, and the nation through a variety of academic, research and service activities.
Our missions are to: holistically educate our students for professional leadership as creative problem-solvers in a diverse society, conduct advanced research for societal advancement, train graduate students for scholarly inquiry, and engage with alumni, industry, government, and community partners through outreach activities. In order to produce engineers prepared for success across a range of career paths, our academic program integrates training in engineering principles, critical thinking, hands-on projects, open-ended problem solving, and the essential skills of teamwork, communication, and ethics.
Browse
Browsing Department of Mechanical Engineering by Subject "0913 Mechanical Engineering"
Now showing 1 - 8 of 8
Results Per Page
Sort Options
- Communicating Inferred Goals With Passive Augmented Reality and Active Haptic FeedbackMullen, James F.; Mosier, Josh; Chakrabarti, Sounak; Chen, Anqi; White, Tyler; Losey, Dylan P. (IEEE, 2021-10-01)Robots learn as they interact with humans. Consider a human teleoperating an assistive robot arm: as the human guides and corrects the arm's motion, the robot gathers information about the human's desired task. But how does the human know what their robot has inferred? Today's approaches often focus on conveying intent: for instance, using legible motions or gestures to indicate what the robot is planning. However, closing the loop on robot inference requires more than just revealing the robot's current policy: the robot should also display the alternatives it thinks are likely, and prompt the human teacher when additional guidance is necessary. In this letter we propose a multimodal approach for communicating robot inference that combines both passive and active feedback. Specifically, we leverage information-rich augmented reality to passively visualize what the robot has inferred, and attention-grabbing haptic wristbands to actively prompt and direct the human's teaching. We apply our system to shared autonomy tasks where the robot must infer the human's goal in real-time. Within this context, we integrate passive and active modalities into a single algorithmic framework that determines when and which type of feedback to provide. Combining both passive and active feedback experimentally outperforms single modality baselines; during an in-person user study, we demonstrate that our integrated approach increases how efficiently humans teach the robot while simultaneously decreasing the amount of time humans spend interacting with the robot. Videos here: https://youtu.be/swq_u4iIP-g
- Design and Preliminary Evaluation of Two Tool Support Arm Exoskeletons for Gravity CompensationHull, Joshua; Turner, Ranger; Asbeck, Alan T. (2022-03)We present two new arm exoskeletons based on a pantograph linkage. The exoskeletons can support a tool or load held in the hand of up to 9 kg. The pantograph linkage follows the geometry of a user’s upper arm and forearm, and supports the held mass with any arm orientation. The downward force of the load weight is supported by a fulcrum near the user’s shoulder and balanced by a countering downward force behind the user’s back. We present a Passive exoskeleton, which uses a gas spring mechanism to supply the downward force, and an Active exoskeleton that uses a motor. We analyze the forces produced by the gas spring mechanism as its geometry is varied, and explore different possible arrangements of the gas spring mechanism or motor relative to the pantograph linkage. We derive the equations governing the pantograph linkage, and simulate the forces resulting from the two exoskeletons. We conduct measurements of the forces produced by each exoskeleton, and compare them with the simulated forces.
- Learning latent actions to control assistive robotsLosey, Dylan P.; Jeon, Hong Jun; Li, Mengxi; Srinivasan, Krishnan; Mandlekar, Ajay; Garg, Animesh; Bohg, Jeannette; Sadigh, Dorsa (Springer, 2021-08-04)Assistive robot arms enable people with disabilities to conduct everyday tasks on their own. These arms are dexterous and high-dimensional; however, the interfaces people must use to control their robots are low-dimensional. Consider teleoperating a 7-DoF robot arm with a 2-DoF joystick. The robot is helping you eat dinner, and currently you want to cut a piece of tofu. Today’s robots assume a pre-defined mapping between joystick inputs and robot actions: in one mode the joystick controls the robot’s motion in the x–y plane, in another mode the joystick controls the robot’s z–yaw motion, and so on. But this mapping misses out on the task you are trying to perform! Ideally, one joystick axis should control how the robot stabs the tofu, and the other axis should control different cutting motions. Our insight is that we can achieve intuitive, user-friendly control of assistive robots by embedding the robot’s high-dimensional actions into low-dimensional and human-controllable latent actions. We divide this process into three parts. First, we explore models for learning latent actions from offline task demonstrations, and formalize the properties that latent actions should satisfy. Next, we combine learned latent actions with autonomous robot assistance to help the user reach and maintain their high-level goals. Finally, we learn a personalized alignment model between joystick inputs and latent actions. We evaluate our resulting approach in four user studies where non-disabled participants reach marshmallows, cook apple pie, cut tofu, and assemble dessert. We then test our approach with two disabled adults who leverage assistive devices on a daily basis.
- Learning reward functions from diverse sources of human feedback: Optimally integrating demonstrations and preferencesBıyık, Erdem; Losey, Dylan P.; Palan, Malayandi; Landolfi, Nicholas C.; Shevchuk, Gleb; Sadigh, Dorsa (SAGE, 2022-01)Reward functions are a common way to specify the objective of a robot. As designing reward functions can be extremely challenging, a more promising approach is to directly learn reward functions from human teachers. Importantly, data from human teachers can be collected either passively or actively in a variety of forms: passive data sources include demonstrations (e.g., kinesthetic guidance), whereas preferences (e.g., comparative rankings) are actively elicited. Prior research has independently applied reward learning to these different data sources. However, there exist many domains where multiple sources are complementary and expressive. Motivated by this general problem, we present a framework to integrate multiple sources of information, which are either passively or actively collected from human users. In particular, we present an algorithm that first utilizes user demonstrations to initialize a belief about the reward function, and then actively probes the user with preference queries to zero-in on their true reward. This algorithm not only enables us combine multiple data sources, but it also informs the robot when it should leverage each type of information. Further, our approach accounts for the human’s ability to provide data: yielding user-friendly preference queries which are also theoretically optimal. Our extensive simulated experiments and user studies on a Fetch mobile manipulator demonstrate the superiority and the usability of our integrated framework.
- Physical interaction as communication: Learning robot objectives online from human correctionsLosey, Dylan P.; Bajcsy, Andrea; O'Malley, Marcia K.; Dragan, Anca D. (SAGE, 2021-10-25)When a robot performs a task next to a human, physical interaction is inevitable: the human might push, pull, twist, or guide the robot. The state of the art treats these interactions as disturbances that the robot should reject or avoid. At best, these robots respond safely while the human interacts; but after the human lets go, these robots simply return to their original behavior. We recognize that physical human–robot interaction (pHRI) is often intentional: the human intervenes on purpose because the robot is not doing the task correctly. In this article, we argue that when pHRI is intentional it is also informative: the robot can leverage interactions to learn how it should complete the rest of its current task even after the person lets go. We formalize pHRI as a dynamical system, where the human has in mind an objective function they want the robot to optimize, but the robot does not get direct access to the parameters of this objective: they are internal to the human. Within our proposed framework human interactions become observations about the true objective. We introduce approximations to learn from and respond to pHRI in real-time. We recognize that not all human corrections are perfect: often users interact with the robot noisily, and so we improve the efficiency of robot learning from pHRI by reducing unintended learning. Finally, we conduct simulations and user studies on a robotic manipulator to compare our proposed approach with the state of the art. Our results indicate that learning from pHRI leads to better task performance and improved human satisfaction.
- A simulation-based comparative study on lateral characteristics of trucks with double and triple trailersChen, Yang; Peterson, Andrew W.; Zhang, Ce; Ahmadian, Mehdi (Inderscience Publishers, 2019-01-01)This paper investigates the lateral stability and manoeuvrability in long combination vehicles (LCVs), namely semi-trucks with 28-ft doubles, 28- ft triples, and 33-ft doubles, using TruckSim. In particular, the likelihood of rollovers, rearward amplification, and off-tracking are analysed among those LCVs using the multi-domain dynamic models developed in TruckSim. The efforts to validate the truck dynamic model against test results are also included. The simulation results show that trucks with triple trailers exhibit a larger rearward amplification, higher likelihood of rollovers, and larger offtracking than trucks with double trailers. Additionally, the results indicate that increasing the trailer length from 28 to 33 feet does not increase the likelihood of rollovers or the rearward amplification. In fact, the longer trailers provide a slight amount of additional roll stability due to their longer wheelbase.
- A statistical evaluation of multiple regression models for contact dynamics in rail vehicles using roller rig dataHosseini, Sayed Mohammad; Radmehr, Ahmad; Ahangarnejad, Arash Hosseinian; Gramacy, Robert B.; Ahmadian, Mehdi (Taylor & Francis, 2022-01-06)A statistical analysis of a large amount of data from experiments conducted on the Virginia Tech-Federal Railroad Administration (VT-FRA) roller rig under various field-emulated conditions is performed to develop multiple regression models for longitudinal and lateral tractions. The experiment-based models are intended to be an alternative to the classical wheel-rail contact models that have been available for decades. The VT-FRA roller rig data is used to develop parametric regression models that efficiently capture the relationship between traction and the combined effects of the influential variables. Single regression models for representing the individual effect of wheel load, creepage, and angle of attack on longitudinal and lateral traction were investigated by the authors in an earlier study. This study extends single regression models to multiple regression models and assesses the interaction among the variables using model selection approaches. The multiple-regression models are then compared with CONTACT, a well-known modelling tool for contact dynamics, in terms of prediction accuracy. The predictions made by both CONTACT and multiple regression models for longitudinal and lateral tractions are in close agreement with the measured data on the VT-FRA roller rig. The multiple regression model, however, offers an algebraic expression that can be solved far more efficiently than a simulation run in CONTACT for a new dynamic condition. The results of the study further indicate that the established multiple regression models are an effective means for studying the effect of multiple parameters such as wheel load, creepage, and angle of attack on longitudinal and lateral tractions. Such data-driven parametric models provide an essential analysis and engineering tool in contact dynamics, just as they have in many other areas of science and engineering.
- Stiff and strong, lightweight bi-material sandwich plate-lattices with enhanced energy absorptionHsieh, Meng-Ting; Ha, Chan Soo; Xu, Zhenpeng; Kim, Seokpum; Wu, H. Felix; Kunc, Vlastimil; Zheng, Xiaoyu (Springer, 2021-08-17)Plate-based lattices are predicted to reach theoretical Hashin–Shtrikman and Suquet upper bounds on stiffness and strength. However, simultaneously attaining high energy absorption in these plate-lattices still remains elusive, which is critical for many structural applications such as shock wave absorber and protective devices. In this work, we present bi-material isotropic cubic + octet sandwich plate-lattices composed of carbon fiber-reinforced polymer (stiff) skins and elastomeric (soft) core. This bi-material configuration enhances their energy absorption capability while retaining stretching-dominated behavior. We investigate their mechanical properties through an analytical model and finite element simulations. Our results show that they achieve enhanced energy absorption approximately 2–2.8 times higher than their homogeneous counterparts while marginally compromising their stiffness and strength. When compared to previously reported materials, these materials achieve superior strength-energy absorption characteristics, making them an excellent candidate for stiff and strong, lightweight energy absorbing applications. Graphic Abstract: [Figure not available: see fulltext.]