Browsing by Author "Nittala, Aditya Shekhar"

Now showing 1 - 4 of 4
  • Thumbnail Image
    ItemOpen Access
    Dynamic Locomotion for Humanoid Robots Via Deep Reinforcement Learning
    (2022-09) Garza Bayardo, Rodrigo Alberto; Ramirez-Serrano, Alejandro; Lee, Jihyun; Nittala, Aditya Shekhar
    A longstanding goal in legged-mobile robotics is to enable robots to learn robust control policies capable of expanding their locomotion abilities to be competent and efficient when moving on a priori unknown uneven terrains. Traditional control techniques lack the generalizability required for legged robots to locomote outside a controlled lab environment in terrains with diverse difficult-to-model uncertainties (e.g., friction coefficients, compliance/deformation, etc.). The approach presented in this thesis combines Deep Reinforcement Learning (Deep RL) with motion capture (MoCap) data to train a biped simulated agent (i.e., humanoid robot) to perform a rich repertoire of diverse skills (e.g., walking, crawling, running, climbing steps, etc.). The method begins by creating reference motion clips with the robot’s morphology for desired locomotion gaits. This is achieved by applying motion retargeting techniques to adapt MoCap clips (taken from humans) to the humanoid system. Subsequently, these adapted reference clips are used as inputs to a new Deep RL architecture. Such architecture uses the Proximal Policy Optimization (PPO) algorithm to train the simulated agent to perform the gait of interest on randomized domains that vary with each iteration. After training, the generated control policy is transferred to a real robot for testing and fine-tuning. The benefits of this approach include i) reducing the training time typically required in multi-legged systems, ii) avoiding exposing the robot and the trainer (human operator) to the tedious and time-consuming initial iterations of the learning process where mistakes are likely to happen, and iii) generalizability – the ability to employ the training model on virtually every available legged mobile robot. Thus, the proposed approach enables robots to locomote in real-world settings. The results are demonstrated through simulation and experimentally tested on a Robotis’ THORMANG 3.0 humanoid robot.
  • Thumbnail Image
    ItemOpen Access
    SHVIL, PLANWELL, & FLYING FRUSTUM: Spatial Interaction With 3D Physical Maps
    (2016-01-29) Nittala, Aditya Shekhar; Sousa, Prof. Mario Costa; Sharlin, Prof. Ehud; Sousa, Prof. Mario Costa; Sharlin, Prof. Ehud; Takashima, Prof. Kazuki; Viczko, Prof. April A.
    Spatial representations are crucial when people interact with the physical environment. For example, geographic maps are one of the primary sources for way-finding, spatial planning and navigational activities. Recent technological advancements enable the evolution of current 2D interactive spatial representations of the maps to physical 3D interactive representations using techniques such as 3D printing and mixed reality interaction. In this thesis, we undertake the task of designing collaborative spatial interaction techniques for physical representation of maps. We designed interfaces for the following application scenarios: collaborative terrain navigation, petroleum-well planning, and remote unmanned aerial vehicle (UAV) control. We present our research, encompassing three prototypes we designed and implemented: Shvil, an augmented reality interface for collaborative terrain navigation; PlanWell, a spatial user interface for collaborative petroleum well planning; and Flying Frustum, a spatial interface for enhancing human-UAV awareness.
  • Thumbnail Image
    ItemOpen Access
    SonicData: Broadcasting Data via Sound for Smartphones
    (2014-10-29) Nittala, Aditya Shekhar; Yang, Xing-Dong; Sharlin, Ehud; Bateman, Scott; Greenberg, Saul
    SonicData is a technique for broadcasting data to smartphones via audio streams using phone’s built-in microphone. SonicData augments an audio stream in the environment with nearly inaudible high-frequencies, allowing data to be sent to any smartphone in the vicinity using regular speakers and without any need for special hardware and software infrastructure or handshaking requirements. We detail the technical implementation of the SonicData prototype, outline a technical evaluation of its capabilities, and describe the results of a preliminary study of its effect on the quality of sound streams. We designed four interaction techniques that highlight SonicData’s potential as a complementary technique for broadcasting data to smartphones.
  • Thumbnail Image
    ItemOpen Access
    Understanding Gesture and Microgesture Inputs for Augmented Reality Maps
    (2024-07-01) Danyluk, Kurtis; Klueber, Simon; Nittala, Aditya Shekhar; Willett, Wesley
    We explore the potential for subtle on-hand gesture and microgesture interactions for map navigation with augmented reality (AR) devices. We describe a design exercise and follow-up elicitation study in which we identified on-hand gestures for cartographic interaction primitives. Microgestures and on-hand interactions are a promising space for AR map navigation as they offers always-available, tactile, and memorable spaces for interaction. Our findings show a clear set of microgesture interaction patterns that are well suited for supporting map navigation and manipulation. In particular, we highlight how the properties of various microgestures align with particular cartographic interaction tasks. We also describe our experience creating an exploratory proof-of-concept AR map prototype which helped us identify new opportunities and practical challenges for microgesture control. Finally, we discuss how future AR map systems could benefit from on-hand and microgesture input schemes.