About

Nisar Ahmed is an Associate Professor in the Ann and H.J. Smead Aerospace Engineering Sciences at the University of Colorado Boulder, where he also serves as the Director of the Research and Engineering Center for Unmanned Vehicles (RECUV). His educational background includes a PhD and MS in Mechanical Engineering with a focus on Dynamics, Systems, and Controls from Cornell University, and a BS in Engineering from The Cooper Union for the Advancement of Science and Art. His research interests encompass collaborative human and autonomous robot vehicle systems, dynamic state estimation and sensor fusion, supervisory control and decentralized coordination in networked systems, as well as the application of statistical system identification, machine learning, and artificial intelligence to aerospace challenges. He has held various roles at CU Boulder, including Site Director for the NSF IUCRC Center for Aerial Autonomy, Mobility, and Sensing, and has been recognized with awards such as the H. Joseph Smead Faculty Fellow and the AIAA Guidance, Navigation and Control Conference Best Paper Award.

Research topics

• Artificial Intelligence
• Computer Science
• Machine Learning
• Mathematics
• Physics
• Human–computer interaction
• Psychology
• Neuroscience
• Geography
• Engineering
• Aerospace engineering
• Cartography
• Simulation
• Geometry

Selected publications

• Competency-Aware Collaborative Robotic Surface Exploration: A Study at the Mars Desert Research Station

  IEEE transactions on field robotics. · 2025-01-01

  article

  Future concepts for space exploration envision astronauts relying on autonomous robots to accomplish a variety of tasks, including planetary surface exploration, science, and mining operations. An astronaut’s understanding of the robots’ competencies will be crucial to informed, risk-aware decision-making during both mission planning and mission execution. A key challenge is to develop methods to calibrate future astronauts (or more generally, users of autonomous robots in safety-critical operations) to the capabilities and limitations of these machines. In the spring of 2024, we partnered with the Mars Society’s Mars Desert Research Station (MDRS) Crew #297 to conduct a two-week exploratory field deployment and limited user study evaluating an autonomous ground robot with a competency-awareness capability—autonomy algorithms capable of quantifying and communicating estimates of the robot’s ability to meet mission objectives. During the analog Mars expedition, the crew utilized a robot to explore areas and gather data during extravehicular activities (EVAs). They overcame several technical challenges and were able to safely and efficiently leverage the robot for their mission needs. The competency information reported by the robot served as a valuable component of the crew’s planning and decision-making process, with 50% of the crew indicating that they directly relied on the assessments while planning tasks for the robot. We believe that integrating competency assessment capabilities into robotic autonomy can enable future astronauts to better manage risk and make more informed decisions while tasking and supervising autonomous robots.

  PublisherDOI

• A multi scale spatial attention based zero shot learning framework for low light image enhancement

  Scientific Reports · 2025-12-03

  articleOpen accessSenior author

  Low-light image enhancement remains a challenging task, particularly in the absence of paired training data. In this study, we present LucentVisionNet, a novel zero-shot learning framework that addresses the limitations of traditional and deep learning-based enhancement methods. The proposed approach integrates multi-scale spatial attention with a deep curve estimation network, enabling fine-grained enhancement while preserving semantic and perceptual fidelity. To further improve generalization, we adopt a recurrent enhancement strategy and optimize the model using a composite loss function comprising six tailored components, including a novel no-reference image quality loss inspired by human visual perception. Extensive experiments on both paired and unpaired benchmark datasets demonstrate that LucentVisionNet consistently outperforms state-of-the-art supervised, unsupervised, and zero-shot methods across multiple full-reference and no-reference image quality metrics. Our framework achieves high visual quality, structural consistency, and computational efficiency, making it well-suited for deployment in real-world applications such as mobile photography, surveillance, and autonomous navigation.

  PublisherDOI

• Rao-Blackwellized POMDP Planning

  2025-05-19

  article

  Partially Observable Markov Decision Processes (POMDPs) provide a structured framework for decision-making under uncertainty, but their application requires efficient belief updates. Sequential Importance Resampling Particle Filters (SIRPF), also known as Bootstrap Particle Filters, are commonly used as belief updaters in large approximate POMDP solvers, but they face challenges such as particle deprivation and high computational costs as the system's state dimension grows. To address these issues, this study introduces Rao-Blackwellized POMDP (RB-POMDP) approximate solvers and outlines generic methods to apply Rao-Blackwellization in both belief updates and online planning. We compare the performance of SIRPF and Rao-Blackwellized Particle Filters (RBPF) in a simulated localization problem where an agent navigates toward a target in a GPS-denied environment using POMCPOW and RB-POMCPOW planners. Our results not only confirm that RBPFs maintain efficient belief approximations over time with fewer particles, but, more surprisingly, RBPFs combined with quadrature-based integration improve planning quality significantly compared to SIRPF-based planning under the same computational limits.

  PublisherDOI

• Scalable Factor Graph-Based Heterogeneous Bayesian DDF for Dynamic Systems

  IEEE Transactions on Robotics · 2025-11-25

  articleSenior author

  Heterogeneous Bayesian decentralized data fusion captures the set of problems in which two or more robots must combine probability density functions over non-equal, but overlapping sets of random variables. In the context of multi-robot dynamic systems, this enables robots to take a “divide and conquer” approach to reason and share data over complementary tasks instead of over the full joint state space. For example, in a target tracking application, this allows robots to track different subsets of targets and share data on only common targets. This paper presents a system by which robots can each use a local factor graph to represent relevant partitions of a complex global joint probability distribution, thus allowing them to avoid reasoning over the entirety of a more complex model and saving communication as well as computation costs. From a theoretical point of view, this paper makes contributions by casting the heterogeneous decentralized fusion problem in terms of factor graphs, analyzing the challenges that arise due to dynamic filtering, and then developing a new conservative filtering algorithm that ensures statistical correctness. From a practical point of view, we show how this system can be used to represent different multi-robot applications and then test it with simulations and hardware experiments to validate and demonstrate its statistical conservativeness, applicability, and robustness to real-world challenges.

  PublisherDOI

• Active Inference for Bandit-Based Autonomous Robotic Exploration With Dynamic Preferences

  IEEE Transactions on Robotics · 2025-01-01

  articleSenior author

  Autonomous selection of optimal options for data collection from multiple alternatives is challenging in uncertain environments. When secondary information about options is accessible, such problems can be framed as contextual multi-armed bandits (CMABs). Neuro-inspired active inference has gained interest for its ability to balance exploration and exploitation using the expected free energy objective function. Unlike previous studies that showed the effectiveness of active inference based strategy for CMABs using synthetic data, this study aims to apply active inference to realistic scenarios, using a simulated mineralogical survey site selection problem. Hyperspectral data from AVIRIS-NG at Cuprite, Nevada, serves as contextual information for predicting outcome probabilities, while geologists' mineral labels represent outcomes. Monte Carlo simulations assess the robustness of active inference against changing expert preferences. Results show active inference requires fewer iterations than standard bandit approaches with real-world noisy and biased data, and performs better when outcome preferences vary online by adapting the selection strategy to align with expert shifts.

  PublisherDOI

• Multi-Robot Motion Planning with Cooperative Localization

  2025-10-19

  article

  We consider the uncertain multi-robot motion planning (MRMP) problem with cooperative localization (CL-MRMP), under both motion and measurement noise, where each robot can act as a sensor for its nearby teammates. We formalize CL-MRMP as a chance-constrained motion planning problem, and propose a safety-guaranteed algorithm that explicitly accounts for robot-robot correlations. Our approach extends a sampling-based planner to solve CL-MRMP while preserving probabilistic completeness. To improve efficiency, we introduce novel biasing techniques. We evaluate our method across diverse benchmarks, demonstrating its effectiveness in generating motion plans, with significant performance gains from biasing strategies.

  PublisherDOI

• Intelligent Decision Support for Target Tracking Analysis and Characterization

  Journal of Aerospace Information Systems · 2025-10-24

  article

  While automation has been increasingly used to process high volumes of satellite remote sensing data, deriving accurate and actionable information from these data streams still requires human analysts. Machine learning algorithms are critical for processing large datasets, whereas well-trained operators are especially effective at synthesizing information from diverse sources and recognizing unique situations that may require different interpretations of the data. Therefore, combining algorithmic strengths with improved operator awareness is critical for reliable and robust data analysis. This research contributes a suite of algorithms that support the visualization, analysis, and characterization of infrared satellite data. The key components of our Collaborative Analyst-Machine Perception system include a probabilistic classifier, a false data filter, a historical track comparison tool, and an online data recommendation system. These components are integrated into an interactive dashboard and trained on synthetic satellite information that emulates operational challenges. We evaluated our application with six United States Space Force satellite operators in a live scenario with simulated real-time data acquisition. We found that the operators rated our application as more usable than current operational systems and that the combined human–machine team was capable of more accurate data characterization than machine learning algorithms alone.

  PublisherDOI

• A Multi-Scale Spatial Attention-Based Zero-Shot Learning Framework for Low-Light Image Enhancement

  Research Square · 2025-06-27

  preprintOpen accessSenior author
  PublisherOA PDFDOI

• Extended Version: Multi-Robot Motion Planning with Cooperative Localization

  ArXiv.org · 2025-04-08

  preprintOpen access

  We consider the uncertain multi-robot motion planning (MRMP) problem with cooperative localization (CL-MRMP), under both motion and measurement noise, where each robot can act as a sensor for its nearby teammates. We formalize CL-MRMP as a chance-constrained motion planning problem, and propose a safety-guaranteed algorithm that explicitly accounts for robot-robot correlations. Our approach extends a sampling-based planner to solve CL-MRMP while preserving probabilistic completeness. To improve efficiency, we introduce novel biasing techniques. We evaluate our method across diverse benchmarks, demonstrating its effectiveness in generating motion plans, with significant performance gains from biasing strategies.

  PublisherOA PDFDOI

• The Synergy of Low-Code/No-Code and AI/ML: Enhancing Intelligent Automation

  World Journal of Engineering and Technology · 2025-01-01

  articleOpen access1st authorCorresponding

  Software development has been revolutionized by low-code and no-code platforms, which make it possible for even non-programmers to create and launch apps rapidly. In contrast to traditional coding, they speed up development with drag-and-drop components, pre-built templates, and ready-to-use plugins. Their effects on accelerating innovation, reducing expenses, and facilitating digital transformation are examined in this article, together with issues including scalability, security, and restricted customization. It also investigates how similar platforms might be used in the future to promote cooperation between technical and non-technical teams. While LCNC platforms help close the gap between IT solutions and business needs, thorough integration is necessary for long-term success.

  PublisherOA PDFDOI

Frequent coauthors

• Mark Campbell

  29 shared

• Alan E. Willner

  University of Southern California

  28 shared

• Yongxiong Ren

  Robert Bosch (Germany)

  28 shared

• Guodong Xie

  24 shared

• Nicholas Conlon

  University of Colorado Boulder

  24 shared

• Eric W. Frew

  22 shared

• Moshe Tur

  Tel Aviv University

  21 shared

• Hao Huang

  21 shared

Education

• Ph.D., Mechanical Engineering (Dynamics, Systems and Controls)

  Cornell University

  2012

• M.S., Mechanical Engineering (Dynamics, Systems and Controls)

  Cornell University

  2009

• B.S.

  The Cooper Union for the Advancement of Science and Art

  2006

Awards & honors

• H. Joseph Smead Faculty Fellow (2021)
• Aerospace Control and Guidance Systems Committee (ACGSC) Dav…
• ASEE Air Force Summer Faculty Fellowship (2014)
• AIAA Guidance, Navigation and Control Conference Best Paper…