Nisar R. Ahmed

Dr. Nisar R. Ahmed

orcid

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 January 2024 in IEEE Transactions on Aerospace and Electronic Systems
Verified
Journal Article
Closed access
ZCZhaozhong Chen
Zhaozhong Chen
HBHarel Biggie
Harel Biggie
NANisar R. Ahmed
Nisar R. Ahmed
SJSimon Julier
Simon Julier
CHChristoffer R. Heckman
Christoffer R. Heckman
The nonlinear and stochastic relationship between noise covariance parameter values and state estimator performance makes optimal filter tuning a very challenging problem. Popular optimization-based tuning approaches can easily get trapped in local minima, leading to poor noise parameter identification and suboptimal state estimation. Recently, black box techniques based on Bayesian optimization with Gaussian processes (GPBO) have been shown to overcome many of these issues, using normalized estimation error squared (NEES) and normalized innovation error (NIS) statistics to derive cost functions for Kalman filter auto-tuning. While reliable noise parameter estimates are obtained in many cases, GPBO solutions obtained with these conventional cost functions do not always converge to optimal filter noise parameters and lack robustness to parameter ambiguities in time-discretized system models. This paper addresses these issues by making two main contributions. First, new cost functions are developed to determine if an estimator has been tuned correctly. It is shown that traditional chi-square tests are inadequate for correct auto-tuning because they do not accurately model the distribution of innovations when the estimator is incorrectly tuned. Second, the new metrics (formulated over multiple time discretization intervals) is combined with a Student-t processes Bayesian Optimization (TPBO) to achieve robust estimator performance for time discretized state space models. The robustness, accuracy, and reliability of our approach are illustrated on classical state estimation problems.

27 August 2019 in IEEE Transactions on Robotics
Verified
Journal Article
Open access
LBLuke Burks
Luke Burks
ILIan Loefgren
Ian Loefgren
NANisar R. Ahmed
Nisar R. Ahmed
This article develops novel strategies for optimal planning with semantic observations using continuous state partially observable Markov decision processes (CPOMDPs). Two major innovations are presented in relation to Gaussian mixture (GM) CPOMDP policy approximation methods. While existing methods have many desirable theoretical properties, they are unable to efficiently represent and reason over hybrid continuous-discrete probabilistic models. The first major innovation is the derivation of closed-form variational Bayes GM approximations of point-based value iteration Bellman policy backups, using softmax models of continuous-discrete semantic observation probabilities. A key benefit of this approach is that dynamic decision making tasks can be performed with complex non-Gaussian uncertainties, while also exploiting continuous dynamic state-space models (thus avoiding cumbersome and costly discretization). The second major innovation is a new clustering-based technique for mixture condensation that scales well to very large GM policy functions and belief functions. Simulation results for a target search and interception task with semantic observations show that the GM policies resulting from these innovations are more effective than those produced by other state-of-the-art policy approximations, but require significantly less modeling overhead and online runtime cost. Additional results show the robustness of this approach to model errors and scaling to higher dimensions.

9 April 2024 in ACM Computing Surveys
Verified
Journal Article
Open access
NCNicholas Conlon
Nicholas Conlon
NANisar R. Ahmed
Nisar R. Ahmed
DSDaniel Szafir
Daniel Szafir
Humans working with autonomous artificially intelligent systems may not be experts in the inner workings of their machine teammates, but need to understand when to employ, trust, and rely on the system. A critical challenge is to develop machine agents with the capacity to understand their own capabilities and limitations, and the ability to communicate this information to human partners. Self-assessment is an emerging field that tackles this challenge through the development of algorithms that enable autonomous agents to understand and communicate their competency. These methods can engender appropriate trust and align human expectations with autonomous assistant abilities. However, current research in self-assessment is dispersed across many fields, including artificial intelligence, robotics, and human factors. This survey connects work from these disparate areas and reviews state-of-the-art methods for algorithmic self-assessments that enable autonomous agents to estimate, understand, and communicate valuable information pertaining to their competency, with focus on methods that can improve interactions within human-machine teams. To better understand the landscape of self-assessment approaches, we present a framework for categorizing work in self-assessment based on underlying algorithm type: test-based , learning-based , or knowledge-based . We synthesize common features across these approaches and discuss relevant future directions for research in this emerging space.

14 December 2022 in IEEE Transactions on Robotics
Verified
Journal Article
Closed access
ODOfer Dagan
Ofer Dagan
NANisar R. Ahmed
Nisar R. Ahmed
In Bayesian peer-to-peer decentralized data fusion, the underlying distributions held locally by autonomous agents are frequently assumed to be over the same set of variables (homogeneous). This requires each agent to process and communicate the full global joint distribution, and thus, leads to high computation and communication costs irrespective of relevancy to specific local objectives. This work formulates and studies heterogeneous decentralized fusion problems, defined as the set of problems in which either the communicated or the processed distributions describe different, but overlapping, random states of interest that are subsets of a larger full global joint state. We exploit the conditional independence structure of such problems and provide a rigorous derivation of novel exact and approximate conditionally factorized heterogeneous fusion rules. We further develop a new version of the homogeneous channel filter algorithm to enable conservative heterogeneous fusion for smoothing and filtering scenarios in dynamic problems. Numerical examples show more than 99.5% potential communication reduction for heterogeneous channel filter fusion, and a multitarget tracking simulation shows that these methods provide consistent estimates while remaining computationally scalable.

19 April 2023 in IEEE Transactions on Robotics
Verified
Journal Article
Closed access
LBLuke Burks
Luke Burks
HRHunter M. Ray
Hunter M. Ray
JMJamison McGinley
Jamison McGinley
SVSousheel Vunnam
Sousheel Vunnam
NANisar R. Ahmed
Nisar R. Ahmed
The ability of autonomous robots to model, communicate, and act on semantic “soft data” remains challenging. The human-assisted robotic planning and sensing (HARPS) framework is presented for active semantic sensing and planning in human–robot teams to address these gaps by formally combining the benefits of online sampling-based partially observable Markov decision process policies, multimodal human–robot interaction, and Bayesian data fusion. HARPS lets humans impose model structure and extend the range of soft data by sketching and labeling new semantic features in uncertain environments. Dynamic model updating lets robotic agents actively query humans for novel and relevant semantic data, thereby improving model and state beliefs for improved online planning. Simulations of a unmanned aerial vehicle-enabled target search in a large-scale partially structured environment show significant improvements in time and beliefs required for interception versus conventional planning with robot-only sensing. A human subject study in the same environment shows an average doubling in dynamic target capture rate compared to the lone robot case and highlights the robustness of HARPS over a range of user characteristics and interaction modalities.

26 July 2018 in IEEE Signal Processing Letters
Verified
Journal Article
Closed access
NANisar R. Ahmed
Nisar R. Ahmed
This note considers softmax parameter estimation when little/no labeled training data is available, but a priori information about the relative geometry of class label log-odds boundaries is available. It is shown that “data-free” softmax model synthesis corresponds to solving a linear system of parameter equations, wherein desired dominant class log-odds boundaries are encoded via convex polytopes that decompose the input feature space. When solvable, the linear equations yield closed-form softmax parameter solution families using class boundary polytope specifications only. This allows softmax parameter learning to be implemented without expensive brute force data sampling and numerical optimization. The linear equations can also be adapted to constrained maximum likelihood estimation in data-sparse settings. Since solutions may also fail to exist for the linear parameter equations derived from certain polytope specifications, it is thus also shown that there exist probabilistic classification problems over m convexly separable classes for which the log-odds boundaries cannot be learned using an m-class softmax model.