About

Jennifer Bird is an Associate Professor of Voice (soprano) at the College of Music, University of Colorado Boulder. She has built an international reputation as a charismatic, intelligent, and versatile performer, having performed more than 50 roles across opera, operetta, and musical theater, as well as standard oratorio, concert, and recital literature. Her background includes studying in Germany at the Hamburg Musikhochschule, where she joined the vocal studio of Judith Beckmann, supported by a Rotary International Ambassadorial Scholarship. Bird has held engagements at numerous prominent theaters and opera houses, including the Landestheater Coburg, Bremer Theater, Vienna Volksoper, Nationaltheater Mannheim, and others, performing major roles such as Lulu, Lucia di Lammermoor, Gilda, Violetta, and Konstanze, among others. She has also been a soprano soloist in notable performances with orchestras and at prestigious venues, including Carnegie Hall, Alice Tully Hall, and the Stuttgart Konzerthalle. Her recent seasons feature performances of works by Mendelssohn, Mozart, Mahler, Poulenc, Bach, Brahms, and others, often collaborating with renowned ensembles and conductors. Bird holds a Bachelor of Music and a Bachelor of Arts from Southern Methodist University and a Master of Music from the University of Michigan. As a faculty member at the University of Colorado Boulder, her students perform widely in leading roles and professional ensembles internationally.

Research topics

• Computer Science
• Computer Security
• Simulation
• Telecommunications
• Physical therapy
• Multimedia
• Nursing
• Environmental science
• Meteorology
• Aerospace engineering
• Intensive care medicine
• Computer network
• Emergency medicine
• Geography
• Engineering
• Surgery
• Medicine
• Systems engineering

Selected publications

• Efficiently Training Observable Control Policies

  2025-01-03

  articleSenior author

  Control policies which enable the agent state to be estimated from observations of the actions chosen hold promise in enabling multi-agent coordination in communication-limited scenarios. When policies are produced by a learning system they can be made observable by directly rewarding estimator performance. This is computationally expensive, requires the estimator to be formulated prior to the control policy training, and closely couples the estimator and policy. As an alternative we investigate rewarding estimator-agnostic measures of observability at training time. Policies are trained against several reward formulations and evaluated using Monte Carlo simulation for a UAS tracking task. These observable policies do not result in as good a state estimate as a policy which directly rewards estimator performance during training, but still achieve substantial improvements over policies which do not consider observability at all while having little impact on the shape of the resulting trajectory.

  PublisherDOI

• Autonomous Uncrewed Aircraft for Mobile Operations in Severe Weather

  Lecture notes in computer science · 2025-08-25

  book-chapter1st authorCorresponding
  PublisherDOI

• Enabling Inter-Vehicle Coordination Through Observable Control Policies

  2024-01-04 · 1 citations

  articleSenior author

  Inter-vehicle coordination in human-autonomy and autonomy-autonomy scenarios is challenging due to limitations in the modality and capacity of aviation communications networks. To integrate advanced autonomous systems equipped with learning components into the national airspace system, it is desirable that the policies guiding these systems are interpretable. Using the ability to estimate the agent’s state from observed control actions as a surrogate for interpretability by both other agents and humans, we aim to generate control policies which are observable. By including an estimator performance term in the cost function used to train control policies, we produce policies which have enhanced observability and evaluate this approach with an aircraft tracking problem. The states of an aircraft executing this observable policy can be more accurately reconstructed from observations of its actions. The performance on the nominal task is not degraded by rewarding observability. This suggests that using observable control policies may provide a means to enable inter-vehicle coordination and to enhance the interpretability of autonomous aircraft in interactions with human airspace users.

  PublisherDOI

• Experimental Assessment of Chance-Constrained Motion Planning for Small Uncrewed Aircraft

  Field Robotics · 2024-01-01 · 3 citations

  articleOpen access

  This work extends the experimental evaluation of chance-constrained motion planning algorithms to fielded fixed-wing small uncrewed aircraft systems (sUAS). Despite advances in planning algorithms, certain challenges remain to producing trajectories for nonholonomic mobile robotic systems such as sUAS. These challenges include nonlinear dynamics which create a complex mapping from inputs to outputs, initialization uncertainty in online motion planning due to compute time of planners and latency in data transfer, environmental uncertainty that has non-Gaussian impact on the robot’s trajectory, and system uncertainty that arises from incomplete models of complex systems. Small UAS often have proprietary components such as commercial, off-the-shelf autopilots which prevent motion planning models from accurately capturing system behavior in all regions of the state space. These challenges can be addressed by leveraging probabilistic motion planners that accurately represent and reason over dynamics and uncertainty. Chance-constrained motion planning offers a method of reasoning over uncertainty as feasibility constraints, and Monte Carlo sampling within a motion planning algorithm offers a method of representing complex uncertainty that may not have a closed form representation. This work extends the chance-constrained motion planning problem to reason over constraint on the trajectory and constraints on the state which ensure that the system model remains valid. Dynamical and systems models are formulated to extend to a broad class of fixed-wing UAS through the experimental derivation of input distributions and system parameters. Uncertainty is quantified using a data-driven approach to modeling input distributions, and Monte Carlo uncertainty sampling deployed within a Rapidly-exploring Random Tree motion planning algorithm is used to plan a trajectory containing a representation of uncertainty. The motion planning algorithm’s ability to accurately reason over layered chance constraints is evaluated experimentally in 61 fielded missions. The results show that when the flight conditions fall within the domain of the uncertainty distributions, the motion planning system is able to accurately reason over the chance constraints.

  PublisherDOI

• An Autonomous Uncrewed Aircraft System Performing Targeted Atmospheric Observation for Cloud Seeding Operations

  Field Robotics · 2023-05-01 · 9 citations

  articleOpen access

  This paper presents the results of a 3-week-long field deployment of an autonomous uncrewed aircraft system for targeted observation of early-convective storm systems in the U.S. Great Plains with application to cloud seeding operations. Due to reduced operational costs and requirements, autonomous small uncrewed aircraft systems present an appealing alternative to traditional crewed aircraft. The objective of the system is to gather and ultimately act upon in situ atmospheric data that are inaccessible via remote sensing techniques. Utilizing a combination of remote and in situ weather data, a dispersed autonomous decision-making system works integrally with a human operator to investigate early-convective storms for subregions which have favorable conditions for cloud seeding. The autonomy framework enables one operator to interface with multiple aircraft, which is demonstrated by performing complex sensing and seeding maneuvers with a team of two aircraft. Results from nine flights totaling over 8 hours of flight time are presented and discussed. Although the release of actual cloud seeding material was not performed during the campaign, this study demonstrates the utility and feasibility of small uncrewed aircraft systems for use in airborne cloud seeding operations.

  PublisherDOI

• Modern and prospective technologies for weather modification activities: A first demonstration of integrating autonomous uncrewed aircraft systems

  Atmospheric Research · 2023 · 24 citations

  • Computer Science
  • Environmental science
  • Computer Science

  We successfully implemented a framework involving an uncrewed aircraft system (UAS) with atmospheric and cloud microphysical sensing and autonomous adaptive control technologies to search, identify, carry out, monitor and/or evaluate cloud seeding operations to enhance precipitation. This first-time implementation featured trials in the U.S. Great Plains area following the US aviation authority regulations during a three-week summer trial period. The trials provide insight into using UAS in an operational setting. Their broader significance is a step toward improved targeting efficiency, and hence improved operational cloud seeding activity effectiveness. The implementation of the integrated UAS provide and exploit the temporal and spatial cloud-scale sensitivities information to overcome the operational and natural uncertainties or sparseness of environmental parameters needed to increase cloud seeding operational efficiency.

  PublisherDOI

• Failure and Reliability Analyses for Long-endurance UAS Architecture

  2023-01-01

  articleOpen accessSenior author

  Uninhabited aircraft systems must be able to successfully complete design missions with a probability of failure deemed acceptable. Developing the architecture necessary for such tasks with only physical testing and prototyping is often too costly and time-consuming to be practical. Thus, it is essential to identify modes of failure, assess the probability of these failures, and mitigate these through design of the aircraft systems architecture. This paper examines the architecture of a long-endurance UAS using fault tree analysis to build a risk assessment matrix, allowing for the acceptability of those failures to be determined.

  PublisherOA PDFDOI

• Estimating System State from the Actions of a Reinforcement Learning Agent

  AIAA SCITECH 2023 Forum · 2023-01-19 · 2 citations

  articleSenior author

  View Video Presentation: https://doi.org/10.2514/6.2023-2657.vid Interaction between robotic agents typically requires some information about the other systems' state be available to each agent. Whether through direct communication among cooperating agents or the use of tracking systems for uncooperative agents, this information is critical to enable one agent to respond to another. When agents are not cooperating, or when environmental or operational considerations prevent communication, this interaction requires that agents maximize the use of sparse channels of information. One such channel is the actions selected by an agent. When these actions are selected by a stochastic policy function this can allow an estimation problem to be formulated that attempts to determine the state distribution of an agent given observations of its actions. This paper examines this estimation problem, formulates estimators which take advantage of an a priori known policy function, and examines the performance of the estimator in two example applications. The estimator performance is analyzed to identify some considerations in agent and estimator design to enable estimating state from the agent actions.

  PublisherDOI

• Monte Carlo Uncertainty Characterization & Chance Constraint Design in Motion Planning for Fielded sUAS

  AIAA SCITECH 2022 Forum · 2022-01-03

  article

  View Video Presentation: https://doi.org/10.2514/6.2022-2547.vid When a small uncrewed aircraft system (sUAS) follows a deterministic motion plan, the resulting aircraft behavior will contain uncertainty due to input uncertainty and changing conditions. In order to generate motion plans for sUAS that satisfy problem constraints concerning the aircraft trajectory, the motion planning system needs to reason over complex and uncertain models. Chance constraints offer a method for reasoning over problem constraints given trajectory uncertainty, and Monte Carlo approximations offer a method for empirically modeling complex uncertainty without requiring simplifications to complex distributions. However, the Monte Carlo approximation of the trajectory distribution will contain error given that a finite number of samples is used, which could lead to an incorrect evaluation of the chance constraint. Additionally, the lack of an analytical representation of the distribution limits the ability to validate whether the approximation represents the true distribution. This work addresses these limitations by exploring buffering chance constraints given approximation error using the Wald interval, and using the Mahalanobis distance to validate whether the Monte Carlo approximation of the trajectory distribution is representative of the true distribution. A motion planning system composed of a sampling-based planning algorithm, a Monte Carlo propagation algorithm, and a chance constraint assessment algorithm is interfaced with a fixed-wing sUAS leveraging dispersed computing to perform simulations and conduct fielded experiments. Results show that the Wald interval is only a sufficient method of buffering the chance constraint given specific distributions, and the state-wise and path-wise Mahalanobis values can be used to tune the system models and validate representative performance.

  PublisherDOI

• An Autopilot Interface to Advance Fixed-Wing UAS Autonomy Research

  AIAA SCITECH 2022 Forum · 2022-01-03 · 1 citations

  article1st authorCorresponding

  View Video Presentation: https://doi.org/10.2514/6.2022-1852.vid Small Uncrewed Aircraft Systems (UAS) often pair autopilot systems with companion computers in research and advanced autonomy applications. The autopilot provides low level, hard-real-time capabilities to run stabilization routines and interface with sensors and actuators. The companion computer is often used for trajectory planning, vision-processing, or other experimental or computationally expensive tasks. Recent improvements in companion computer-autopilot interfaces allow the autopilot to participant fully in robotics message passing systems. Existing interfaces modes are often optimized around multicopter platforms, presenting a challenge to fixed-wing aircraft research. We present hardware-in-the-loop, and flight test results for modifications made to an open source autopilot firmware that enhance interface modes for fixed-wing aircraft research. We also present bench test results for modifications that enable advanced sensor fusion research by isolating research and flight critical estimator systems. In all cases the interfaces allow for control to be quickly regained in the event of a failure of the research system. The implementation allows the research components to be isolated by manual or automatic contingency response systems.

  PublisherDOI

Frequent coauthors

• Eric W. Frew

  17 shared

• Jack W. Langelaan

  Pennsylvania State University

  9 shared

• Katherine Glasheen

  University of Colorado Boulder

  7 shared

• Kevin Kochersberger

  Virginia Tech

  7 shared

• C. Alexander Hirst

  6 shared

• Simon Stanworth

  University of Oxford

  5 shared

• Steve Borenstein

  Cooperative Institute for Research in Environmental Sciences

  4 shared

• Adam Sharkasi

  3 shared

Education

• M.S., Opernklasse

  Hamburg Musikhochschule

Awards & honors

• Audience Favorite Prize (twice) at Landestheater Coburg