Matching Items (36)
Filtering by
- All Subjects: Mechanical Engineering
- Creators: Berman, Spring
Description
Autonomous vehicle technology has been evolving for years since the Automated Highway System Project. However, this technology has been under increased scrutiny ever since an autonomous vehicle killed Elaine Herzberg, who was crossing the street in Tempe, Arizona in March 2018. Recent tests of autonomous vehicles on public roads have faced opposition from nearby residents. Before these vehicles are widely deployed, it is imperative that the general public trusts them. For this, the vehicles must be able to identify objects in their surroundings and demonstrate the ability to follow traffic rules while making decisions with human-like moral integrity when confronted with an ethical dilemma, such as an unavoidable crash that will injure either a pedestrian or the passenger.
Testing autonomous vehicles in real-world scenarios would pose a threat to people and property alike. A safe alternative is to simulate these scenarios and test to ensure that the resulting programs can work in real-world scenarios. Moreover, in order to detect a moral dilemma situation quickly, the vehicle should be able to identify objects in real-time while driving. Toward this end, this thesis investigates the use of cross-platform training for neural networks that perform visual identification of common objects in driving scenarios. Here, the object detection algorithm Faster R-CNN is used. The hypothesis is that it is possible to train a neural network model to detect objects from two different domains, simulated or physical, using transfer learning. As a proof of concept, an object detection model is trained on image datasets extracted from CARLA, a virtual driving environment, via transfer learning. After bringing the total loss factor to 0.4, the model is evaluated with an IoU metric. It is determined that the model has a precision of 100% and 75% for vehicles and traffic lights respectively. The recall is found to be 84.62% and 75% for the same. It is also shown that this model can detect the same classes of objects from other virtual environments and real-world images. Further modifications to the algorithm that may be required to improve performance are discussed as future work.
Testing autonomous vehicles in real-world scenarios would pose a threat to people and property alike. A safe alternative is to simulate these scenarios and test to ensure that the resulting programs can work in real-world scenarios. Moreover, in order to detect a moral dilemma situation quickly, the vehicle should be able to identify objects in real-time while driving. Toward this end, this thesis investigates the use of cross-platform training for neural networks that perform visual identification of common objects in driving scenarios. Here, the object detection algorithm Faster R-CNN is used. The hypothesis is that it is possible to train a neural network model to detect objects from two different domains, simulated or physical, using transfer learning. As a proof of concept, an object detection model is trained on image datasets extracted from CARLA, a virtual driving environment, via transfer learning. After bringing the total loss factor to 0.4, the model is evaluated with an IoU metric. It is determined that the model has a precision of 100% and 75% for vehicles and traffic lights respectively. The recall is found to be 84.62% and 75% for the same. It is also shown that this model can detect the same classes of objects from other virtual environments and real-world images. Further modifications to the algorithm that may be required to improve performance are discussed as future work.
ContributorsSankaramangalam Ulhas, Sangeet (Author) / Berman, Spring (Thesis advisor) / Johnson, Kathryn (Committee member) / Yong, Sze Zheng (Committee member) / Arizona State University (Publisher)
Created2019
Description
One potential application of multi-robot systems is collective transport, a task in which multiple mobile robots collaboratively transport a payload that is too large or heavy to be carried by a single robot. Numerous control schemes have been proposed for collective transport in environments where robots can localize themselves (e.g., using GPS) and communicate with one another, have information about the payload's geometric and dynamical properties, and follow predefined robot and/or payload trajectories. However, these approaches cannot be applied in uncertain environments where robots do not have reliable communication and GPS and lack information about the payload. These conditions characterize a variety of applications, including construction, mining, assembly in space and underwater, search-and-rescue, and disaster response.
Toward this end, this thesis presents decentralized control strategies for collective transport by robots that regulate their actions using only their local sensor measurements and minimal prior information. These strategies can be implemented on robots that have limited or absent localization capabilities, do not explicitly exchange information, and are not assigned predefined trajectories. The controllers are developed for collective transport over planar surfaces, but can be extended to three-dimensional environments.
This thesis addresses the above problem for two control objectives. First, decentralized controllers are proposed for velocity control of collective transport, in which the robots must transport a payload at a constant velocity through an unbounded domain that may contain strictly convex obstacles. The robots are provided only with the target transport velocity, and they do not have global localization or prior information about any obstacles in the environment. Second, decentralized controllers are proposed for position control of collective transport, in which the robots must transport a payload to a target position through a bounded or unbounded domain that may contain convex obstacles. The robots are subject to the same constraints as in the velocity control scenario, except that they are assumed to have global localization. Theoretical guarantees for successful execution of the task are derived using techniques from nonlinear control theory, and it is shown through simulations and physical robot experiments that the transport objectives are achieved with the proposed controllers.
Toward this end, this thesis presents decentralized control strategies for collective transport by robots that regulate their actions using only their local sensor measurements and minimal prior information. These strategies can be implemented on robots that have limited or absent localization capabilities, do not explicitly exchange information, and are not assigned predefined trajectories. The controllers are developed for collective transport over planar surfaces, but can be extended to three-dimensional environments.
This thesis addresses the above problem for two control objectives. First, decentralized controllers are proposed for velocity control of collective transport, in which the robots must transport a payload at a constant velocity through an unbounded domain that may contain strictly convex obstacles. The robots are provided only with the target transport velocity, and they do not have global localization or prior information about any obstacles in the environment. Second, decentralized controllers are proposed for position control of collective transport, in which the robots must transport a payload to a target position through a bounded or unbounded domain that may contain convex obstacles. The robots are subject to the same constraints as in the velocity control scenario, except that they are assumed to have global localization. Theoretical guarantees for successful execution of the task are derived using techniques from nonlinear control theory, and it is shown through simulations and physical robot experiments that the transport objectives are achieved with the proposed controllers.
ContributorsFarivarnejad, Hamed (Author) / Berman, Spring (Thesis advisor) / Mignolet, Marc (Committee member) / Tsakalis, Konstantinos (Committee member) / Artemiadis, Panagiotis (Committee member) / Gil, Stephanie (Committee member) / Arizona State University (Publisher)
Created2020
Description
In this thesis, the problem of designing model discrimination algorithms for unknown nonlinear systems is considered, where only raw experimental data of the system is available. This kind of model discrimination techniques finds one of its application in the estimation of the system or intent models under consideration, where all incompatible models are invalidated using new data that is available at run time. The proposed steps to reach the end goal of the algorithm for intention estimation involves two steps: First, using available experimental data of system trajectories, optimization-based techniques are used to over-approximate/abstract the dynamics of the system by constructing an upper and lower function which encapsulates/frames the true unknown system dynamics. This over-approximation is a conservative preservation of the dynamics of the system, in a way that ensures that any model which is invalidated against this approximation is guaranteed to be invalidated with the actual model of the system. The next step involves the use of optimization-based techniques to investigate the distinguishability of pairs of abstraction/approximated models using an algorithm for 'T-Distinguishability', which gives a finite horizon time 'T', within which the pair of models are guaranteed to be distinguished, and to eliminate incompatible models at run time using a 'Model Invalidation' algorithm. Furthermore, due the large amount of data under consideration, some computation-aware improvements were proposed for the processing of the raw data and the abstraction and distinguishability algorithms.The effectiveness of the above-mentioned algorithms is demonstrated using two examples. The first uses the data collected from the artificial simulation of a swarm of agents, also known as 'Boids', that move in certain patterns/formations, while the second example uses the 'HighD' dataset of naturalistic trajectories recorded on German Highways for vehicle intention estimation.
ContributorsBhagwat, Mohit Mukul (Author) / Yong, Sze Zheng (Thesis advisor) / Berman, Spring (Committee member) / Xu, Zhe (Committee member) / Arizona State University (Publisher)
Created2021
Description
This paper introduces a variable impedance controller which dynamically modulates both its damping and stiffness to improve the trade-off between stability and agility in coupled human-robot systems and reduce the human user’s effort. The controller applies a range of robotic damping from negative to positive values to either inject or dissipate energy based on the user’s intent of motion. The controller also estimates the user’s intent of direction and applies a variable stiffness torque to stabilize the user towards an estimated ideal trajectory. To evaluate the controller’s ability to improve the stability/agility trade-off and reduce human effort, a study was designed for human subjects to perform a 2D target reaching task while coupled with a wearable ankle robot. A constant impedance condition was selected as a control with which to compare the variable impedance condition. The position, speed, and muscle activation responses were used to quantify the user’s stability, agility, and effort, respectively. Stability was quantified spatially and temporally, with both overshoot and stabilization time showing no statistically significant difference between the two experimental conditions. Agility was quantified using mean and maximum speed, with both increasing from the constant impedance to variable impedance condition by 29.8% and 59.9%, respectively. Effort was quantified by the overall and maximum muscle activation data, both of which showed a ~10% reduction in effort. Overall, the study demonstrated the effectiveness of the variable impedance controller.
ContributorsArnold, James (Author) / Lee, Hyunglae (Thesis advisor) / Berman, Spring (Committee member) / Yong, Sze Zheng (Committee member) / Arizona State University (Publisher)
Created2021
Description
Many medical procedures, like surgeries, deal with the physical manipulation of sensitive internal tissues. Over time, new medical tools and techniques have been developed to improve the safety and efficacy of these procedures. Despite the leaps and bounds of progress made up to the present day, three major obstacles (among others) persist, bleeding, pain, and the risk of infection. Advances in minimally invasive treatments have transformed many formerly risky surgical procedures into very safe and highly successful routines. Minimally invasive surgeries are characterized by small incision profiles compared to the large incisions in open surgeries, minimizing the aforementioned issues. Minimally invasive procedures lead to several benefits, such as shorter recovery time, fewer complications, and less postoperative pain. In minimally invasive surgery, doctors use various techniques to operate with less damage to the body than open surgery. Today, these procedures have an established, successful history and promising future. Steerable needles are one of the tools proposed for minimally invasive operations. Needle steering is a method for guiding a long, flexible needle through curved paths to reach targets deep in the body, eliminating the need for large incisions. In this dissertation, we present a new needle steering technology: magnetic needle steering. This technology is proposed to address the limitations of conventional needle steering that hindered its clinical applications. Magnetic needle steering eliminates excessive tissue damage, restrictions of the minimum radius of curvature, and the need for a complex nonlinear model, to name a few. It also allows fabricating the needle shaft out of soft and tissue-compliant materials.
This is achieved by first developing an electromagnetic coil system capable of producing desired magnetic fields and gradients; then, a magnetically actuated needle is designed, and its effectiveness is experimentally evaluated. Afterward, the scalability of this technique was tested using permanent magnets controlled with a robotic arm.
Furthermore, different configurations of permanent magnets and their influence on the magnetic field are investigated, enabling the possibility of designing a desired magnetic field for a specific surgical procedure and operation on a particular organ. Finally, potential future directions towards animal studies and clinical trials are discussed.
ContributorsIlami, Mahdi (Author) / Marvi, Hamid (Thesis advisor) / Berman, Spring (Committee member) / Lee, Hyunglae (Committee member) / Nikkhah, Mehdi (Committee member) / Sugar, Thomas (Committee member) / Arizona State University (Publisher)
Created2021
Description
Unmanned aerial vehicles have received increased attention in the last decade due to their versatility, as well as the availability of inexpensive sensors (e.g. GPS, IMU) for their navigation and control. Multirotor vehicles, specifically quadrotors, have formed a fast growing field in robotics, with the range of applications spanning from surveil- lance and reconnaissance to agriculture and large area mapping. Although in most applications single quadrotors are used, there is an increasing interest in architectures controlling multiple quadrotors executing a collaborative task. This thesis introduces a new concept of control involving more than one quadrotors, according to which two quadrotors can be physically coupled in mid-flight. This concept equips the quadro- tors with new capabilities, e.g. increased payload or pursuit and capturing of other quadrotors. A comprehensive simulation of the approach is built to simulate coupled quadrotors. The dynamics and modeling of the coupled system is presented together with a discussion regarding the coupling mechanism, impact modeling and additional considerations that have been investigated. Simulation results are presented for cases of static coupling as well as enemy quadrotor pursuit and capture, together with an analysis of control methodology and gain tuning. Practical implementations are introduced as results show the feasibility of this design.
ContributorsLarsson, Daniel (Author) / Artemiadis, Panagiotis (Thesis advisor) / Marvi, Hamidreza (Committee member) / Berman, Spring (Committee member) / Arizona State University (Publisher)
Created2016