About

Nils Napp is an Assistant Professor in the School of Electrical and Computer Engineering at Cornell University, located in Rhodes Hall, Room 334. His research focuses on design and control strategies for systems that operate with uncertainty, drawing inspiration from evolved biological systems that function reliably in cluttered, unstructured, and fluctuating environments without relying on global information, planning, or communication. His work emphasizes principles such as self-organization, managing noise created by interacting components, and utilizing distributed reactive behaviors as feedback to adapt strategies. In his group, these biological guiding principles are applied to develop algorithms and build robots capable of operating reliably in messy and unstructured real-world environments.

Research topics

• Computer science
• Artificial intelligence
• Distributed computing
• Theoretical computer science
• Computer vision

Selected publications

• Improved Bag-of-Words Image Retrieval with Geometric Constraints for Ground Texture Localization

  arXiv (Cornell University) · 2025-05-16

  preprintOpen accessSenior author

  Ground texture localization using a downward-facing camera offers a low-cost, high-precision localization solution that is robust to dynamic environments and requires no environmental modification. We present a significantly improved bag-of-words (BoW) image retrieval system for ground texture localization, achieving substantially higher accuracy for global localization and higher precision and recall for loop closure detection in SLAM. Our approach leverages an approximate $k$-means (AKM) vocabulary with soft assignment, and exploits the consistent orientation and constant scale constraints inherent to ground texture localization. Identifying the different needs of global localization vs. loop closure detection for SLAM, we present both high-accuracy and high-speed versions of our algorithm. We test the effect of each of our proposed improvements through an ablation study and demonstrate our method's effectiveness for both global localization and loop closure detection. With numerous ground texture localization systems already using BoW, our method can readily replace other generic BoW systems in their pipeline and immediately improve their results.

  PublisherDOI

• Improved Bag-of-Words Image Retrieval with Geometric Constraints for Ground Texture Localization

  2025-05-19

  articleOpen accessSenior author

  Ground texture localization using a downward-facing camera offers a low-cost, high-precision localization solution that is robust to dynamic environments and requires no environmental modification. We present a significantly improved bag-of-words (BoW) image retrieval system for ground texture localization, achieving substantially higher accuracy for global localization and higher precision and recall for loop closure detection in SLAM. Our approach leverages an approximate <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$k$</tex>-means (AKM) vocabulary with soft assignment, and exploits the consistent orientation and constant scale constraints inherent to ground texture localization. Identifying the different needs of global localization vs. loop closure detection for SLAM, we present both high-accuracy and high-speed versions of our algorithm. We test the effect of each of our proposed improvements through an ablation study and demonstrate our method's effectiveness for both global localization and loop closure detection. With numerous ground texture localization systems already using BoW, our method can readily replace other generic BoW systems in their pipeline and immediately improve their results.

  PublisherOA PDFDOI

• Monotone Subsystem Decomposition for Efficient Multi-Objective Robot Design

  2025-05-19

  articleOpen accessSenior author

  Automating design minimizes errors, accelerates the design process, and reduces cost. However, automating robot design is challenging due to recursive constraints, multiple design objectives, and cross-domain design complexity possibly spanning multiple abstraction layers. Here we look at the problem of component selection, a combinatorial optimization problem in which a designer, given a robot model, must select compatible components from an extensive catalog. The goal is to satisfy high-level task specifications while optimally balancing trade-offs between competing design objectives. In this paper, we extend our previous constraint programming approach to multi-objective design problems and propose the novel technique of monotone subsystem decomposition to efficiently compute a Pareto front of solutions for large-scale problems. We prove that subsystems can be optimized for their Pareto fronts and, under certain conditions, these results can be used to determine a globally optimal Pareto front. Furthermore, subsystems serve as an intuitive design abstraction and can be reused across various design problems. Using an example quadcopter design problem, we compare our method to a linear programming approach and demonstrate our method scales better for large catalogs, solving a multi-objective problem of 10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">25</sup> component combinations in seconds. We then expand the original problem and solve a task-oriented, multi-objective design problem to build a fleet of quadcopters to deliver packages. We compute a Pareto front of solutions in seconds where each solution contains an optimal component-level design and an optimal package delivery schedule for each quadcopter.

  PublisherOA PDFDOI

• Robotic Dry-Stacking of Clocháin with Irregular Stones

  2025-05-19

  articleSenior author

  This paper explores automated robotic construction of clocháin<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>, a type of corbelled rock shelter, traditionally crafted by skilled workers. While robots have been employed for simple dry-stacking tasks in the past, such as construction of stone walls or vertical stone towers, the question of whether robots possess the capacity to construct more functional structures remains unanswered. This study presents a significant step forward in robotic dry-stacking of functional structures: the assembly of natural stones into freestanding clocháin structures. We also present a set of stackability measures to aid stone selection, which significantly improves the stability of the planned structures. Our sequential filtering approach, originally designed for planning stone walls, plays a foundational role in achieving stable clochán construction. Experimental results validate the effectiveness of the stackability measures and demonstrate the physical execution of dry-stacking clocháin, The progress demonstrated in this paper opens the door to robotic construction of a wide range of utility structures in unstructured environments.

  PublisherDOI

• Robust Robotic Assembly of Reusable, Rectangular Blocks

  2025-10-19

  articleSenior author

  This paper investigates the importance and design implications for use of rectangular blocks in collective robotic construction systems with distributed control. Specifically, we introduce an automated solver for optimizing the overlaps in user-specified structures; a new robot design capable of manipulating, fastening, and climbing over blocks as wide as the robot; detailed analysis of robot primitives and demonstration of rectilinear, curved, cantilever, and corbeled arch structures; and results from a physics simulator showing how overlaps improve structural integrity when the depositions are noisy. This work represents an important step towards efficient and versatile large-scale robotic construction.

  PublisherDOI

• Monotone Subsystem Decomposition for Efficient Multi-Objective Robot Design

  arXiv (Cornell University) · 2025-05-16

  preprintOpen accessSenior author

  Automating design minimizes errors, accelerates the design process, and reduces cost. However, automating robot design is challenging due to recursive constraints, multiple design objectives, and cross-domain design complexity possibly spanning multiple abstraction layers. Here we look at the problem of component selection, a combinatorial optimization problem in which a designer, given a robot model, must select compatible components from an extensive catalog. The goal is to satisfy high-level task specifications while optimally balancing trade-offs between competing design objectives. In this paper, we extend our previous constraint programming approach to multi-objective design problems and propose the novel technique of monotone subsystem decomposition to efficiently compute a Pareto front of solutions for large-scale problems. We prove that subsystems can be optimized for their Pareto fronts and, under certain conditions, these results can be used to determine a globally optimal Pareto front. Furthermore, subsystems serve as an intuitive design abstraction and can be reused across various design problems. Using an example quadcopter design problem, we compare our method to a linear programming approach and demonstrate our method scales better for large catalogs, solving a multi-objective problem of 10^25 component combinations in seconds. We then expand the original problem and solve a task-oriented, multi-objective design problem to build a fleet of quadcopters to deliver packages. We compute a Pareto front of solutions in seconds where each solution contains an optimal component-level design and an optimal package delivery schedule for each quadcopter.

  PublisherDOI

• Frozen Assets: Leveraging Ice, Water, and Phase Transitions in Robots

  2024-10-14 · 1 citations

  article

  Robots are especially useful in cold, remote, and inhospitable environments such as polar regions and extraterrestrial settings. Due to subfreezing temperatures and limited resources in these environments, robots made of ice are particularly advantageous. In this paper we demonstrate how the solid and liquid phases of water, and transitions between these phases, can be leveraged into common robot designs for modular robots, robot arms, rovers, and soft robots. We explore how robots can utilize structural elements made of ice and exploit the phase change between ice and water to augment their capabilities. Additionally, we do a scaling analysis of ice structural elements to provide insight on their performance at different length scales and ambient temperatures.

  PublisherDOI

• Real-Time Control of Exogenous Carbon Dosing in a Denitrifying Woodchip Bioreactor Treating Agricultural Drainage

  ACS ES&T Engineering · 2024-03-22 · 7 citations

  article

  Nature-based treatment technologies such as denitrifying woodchip bioreactors (WBRs) are employed to manage nitrogen (N) pollution from agricultural nonpoint sources. Due to variability in environmental conditions like temperature and discharge, it is challenging to achieve consistent treatment effectiveness with these passive systems. To improve nitrate (NO3–) load reductions in a field-scale WBR in New York State during cool spring weather, we designed a system for controlled exogenous carbon (C) dosing, allowing rates of C dosing to respond in real time to changing discharge and NO3– concentrations. Treatment efficiencies for NO3–, acetate mass balances, and other bioreactor properties were monitored from April 5 to June 10, 2023. Biostimulation with 7.5 mg C/L acetate (assuming complete mixing of injected acetate with bioreactor pore water) increased NO3– removal rates up to 5-fold compared to a model-based scenario of baseline bioreactor performance, and were as high as 0.4 mg NO3––N L–1 h–1 while water temperatures were <12 °C. Increasing acetate concentrations beyond 7.5 mg C/L did not confer a clear improvement in NO3– removal rates. Cumulative N load reductions increased from 11.3% under the baseline scenario without C dosing to 24.1% with C dosing. The mass ratio of metabolized C to additional N removal was 2.5:1, although the total dosed C/N mass ratio was 5.1:1 due to incomplete acetate utilization in the reactor. We found evidence that C dosing could enhance the future release of dissolved organic N (DON) and dissolved organic C related to biofilm sloughing. The expense of acetate, with a cost efficiency of 86 USD/kg N, was the main cost driver of the real-time control approach. Our results demonstrate the potential of real-time control of C dosing to meaningfully improve nonpoint source N removal during cool spring conditions but also highlight opportunities for methods to improve acetate utilization efficiencies in order to improve the overall cost-effectiveness of the approach.

  PublisherDOI

• Inexpensive, Automated Pruning Weight Estimation in Vineyards

  2024-05-13 · 1 citations

  article

  Pruning weight is indicative of a vine’s ability to produce a crop the following year, informing vineyard management. Current methods for estimating pruning weight are costly, laborious, and/or require specialized know-how and equipment. In this paper we demonstrate an affordable, simple, computer vision-based method to measure pruning weight using a smartphone camera and structured light which produces results better than state-of-the-art techniques for vertical shoot position (VSP) vines and demonstrate initial steps towards estimating pruning weight in high cordon procumbent (HC) vines such as Concord. The simplicity and affordability of this technique lends its self to deployment by farmers today or on future viticulture robotics platforms. We achieved an R2=.80 for VSP vines (better than state-of-the-art computer vision-based methods) and R2=.29 for HC vines (not previously attempted with computer vision-based methods).

  PublisherDOI

• Lunar Infrastructure via Multiscale Granular Stacking

  Journal of Spacecraft and Rockets · 2024-06-10 · 1 citations

  articleSenior author

  Covers advancements in spacecraft and tactical and strategic missile systems, including subsystem design and application, mission design and analysis, materials and structures, developments in space sciences, space processing and manufacturing, space operations, and applications of space technologies to other fields.

  PublisherDOI

Recent grants

• CAREER: Abstraction Barriers for Embodied Algorithms

  NSF · $419k · 2019–2021

Frequent coauthors

• Kirstin Petersen

  Cornell University

  13 shared

• Eric Klavins

  University of Washington

  13 shared

• Vivek Thangavelu

  9 shared

• Radhika Nagpal

  Princeton University

  8 shared

• Frank Schoeneman

  University at Buffalo, State University of New York

  6 shared

• Michael L. Smith

  Auburn University

  6 shared

• Varun Chandola

  6 shared

• Samuel A. Burden

  University of Washington

  6 shared

Awards & honors

• Ten assistant professors win NSF early-career awards