Publications

Refine Results

(Filters Applied) Clear All

Learning emergent discrete message communication for cooperative reinforcement learning

February 24, 2021
  |
Conference Paper
Author:
Sheng Li
…
Published in:
37th Conf. on Uncertainty in Artificial Intelligence, UAI 2021, early access, 26-30 July 2021.
R&D area:
R&D group:

Summary

Communication is a important factor that enables agents work cooperatively in multi-agent reinforcement learning (MARL). Most previous work uses continuous message communication whose high representational capacity comes at the expense of interpretability. Allowing agents to learn their own discrete message communication protocol emerged from a variety of domains can increase the interpretability for human designers and other agents. This paper proposes a method to generate discrete messages analogous to human languages, and achieve communication by a broadcast-and-listen mechanism based on self-attention. We show that discrete message communication has performance comparable to continuous message communication but with much a much smaller vocabulary size. Furthermore, we propose an approach that allows humans to interactively send discrete messages to agents.
READ LESS

Summary

Communication is a important factor that enables agents work cooperatively in multi-agent reinforcement learning (MARL). Most previous work uses continuous message communication whose high representational capacity comes at the expense of interpretability. Allowing agents to learn their own discrete message communication protocol emerged from a variety of domains can increase...

READ MORE
Learning emergent discrete message communication for cooperative reinforcement learning

Towards a distributed framework for multi-agent reinforcement learning research

September 22, 2020
  |
Conference Paper
Author:
Yutai Zhou
…
Published in:
2020 IEEE High Performance Extreme Computing Conf., HPEC, 22-24 September 2020.
Topic:
machine learning
R&D area:
R&D group:

Summary

Some of the most important publications in deep reinforcement learning over the last few years have been fueled by access to massive amounts of computation through large scale distributed systems. The success of these approaches in achieving human-expert level performance on several complex video-game environments has motivated further exploration into the limits of these approaches as computation increases. In this paper, we present a distributed RL training framework designed for super computing infrastructures such as the MIT SuperCloud. We review a collection of challenging learning environments—such as Google Research Football, StarCraft II, and Multi-Agent Mujoco— which are at the frontier of reinforcement learning research. We provide results on these environments that illustrate the current state of the field on these problems. Finally, we also quantify and discuss the computational requirements needed for performing RL research by enumerating all experiments performed on these environments.
READ LESS

Summary

Some of the most important publications in deep reinforcement learning over the last few years have been fueled by access to massive amounts of computation through large scale distributed systems. The success of these approaches in achieving human-expert level performance on several complex video-game environments has motivated further exploration into...

READ MORE
Towards a distributed framework for multi-agent reinforcement learning research

Feature forwarding for efficient single image dehazing

June 17, 2019
  |
Conference Paper
Author:
Peter M. Morales
…
Published in:
IEEE/CVF Conf. on Computer Vision and Pattern Recognition Workshops, CVPRW, 16-17 June 2019.
Topic:
image processing
R&D area:
R&D group:

Summary

Haze degrades content and obscures information of images, which can negatively impact vision-based decision-making in real-time systems. In this paper, we propose an efficient fully convolutional neural network (CNN) image dehazing method designed to run on edge graphical processing units (GPUs). We utilize three variants of our architecture to explore the dependency of dehazed image quality on parameter count and model design. The first two variants presented, a small and big version, make use of a single efficient encoder–decoder convolutional feature extractor. The final variant utilizes a pair of encoder-decoders for atmospheric light and transmission map estimation. Each variant ends with an image refinement pyramid pooling network to form the final dehazed image. For the big variant of the single-encoder network, we demonstrate state-of-the-art performance on the NYU Depth dataset. For the small variant, we maintain competitive performance on the superresolution O/I-HAZE datasets without the need for image cropping. Finally, we examine some challenges presented by the Dense-Haze dataset when leveraging CNN architectures for dehazing of dense haze imagery and examine the impact of loss function selection on image quality. Benchmarks are included to show the feasibility of introducing this approach into real-time systems.
READ LESS

Summary

Haze degrades content and obscures information of images, which can negatively impact vision-based decision-making in real-time systems. In this paper, we propose an efficient fully convolutional neural network (CNN) image dehazing method designed to run on edge graphical processing units (GPUs). We utilize three variants of our architecture to explore...

READ MORE
Feature forwarding for efficient single image dehazing

Geospatial analysis based on GIS integrated with LADAR

October 7, 2013
  |
Journal Article
Author:
Matthew R Fetterman
…
Published in:
Opt. Express, Vol. 21, No. 20, 7 October 2013, pp. 23579-91.
Topic:
ladar
R&D area:
R&D group:

Summary

In this work, we describe multi-layered analyses of a high-resolution broad-area LADAR data set in support of expeditionary activities. High-level features are extracted from the LADAR data, such as the presence and location of buildings and cars, and then these features are used to populate a GIS (geographic information system) tool. We also apply line-of-sight (LOS) analysis to develop a path-planning module. Finally, visualization is addressed and enhanced with a gesture-based control system that allows the user to navigate through the enhanced data set in a virtual immersive experience. This work has operational applications including military, security, disaster relief, and task-based robotic path planning.
READ LESS

Summary

In this work, we describe multi-layered analyses of a high-resolution broad-area LADAR data set in support of expeditionary activities. High-level features are extracted from the LADAR data, such as the presence and location of buildings and cars, and then these features are used to populate a GIS (geographic information system)...

READ MORE
Geospatial analysis based on GIS integrated with LADAR

Collision avoidance for general aviation

October 16, 2011
  |
Conference Paper
Author:
Thomas B. Billingsley
…
Published in:
30th AIAA/IEEE Digital Avionics Systems Conf., 16-20 October 2011.
Topic:
collision avoidance
R&D area:
R&D group:

Summary

The Traffic Alert and Collision Avoidance System (TCAS) is mandated on all large transport aircraft to reduce mid-air collision risk. Since its introduction, no mid-air collisions between TCAS-equipped aircraft have occurred in the United States. However, General Aviation (GA) aircraft are generally not equipped with TCAS and experience collisions several times per year. There is interest in low-cost collision avoidance systems for GA aircraft to reduce collision risk with other GA aircraft as well as with TCAS-equipped aircraft. Since TCAS was designed for large aircraft that can achieve greater vertical rates, the assumptions made by the system and the associated advisories are not always appropriate for GA aircraft. Modifying the TCAS logic to accommodate GA aircraft is far from straightforward. Even minor changes to TCAS to correct operational issues are difficult to implement due to the interaction of the complex rules defining the logic. Recent work has explored an alternative to the TCAS logic based on optimization with respect to a probabilistic model of aircraft behavior. The model encodes performance constraints of GA aircraft, and a computational technique called dynamic programming allows the optimal collision avoidance strategy to be computed efficiently. Prior work has focused on systems that meet the performance assumptions of the existing TCAS logic. However, these assumptions are not always appropriate for GA aircraft. This paper will present simulation results comparing the existing logic to logic that has been optimized to operate onboard GA aircraft. If both aircraft are equipped with collision avoidance logic, it is important that the advisories be coordinated to prevent both aircraft from climbing or descending. The TCAS logic has a built-in coordination mechanism with which a GA system must maintain compatibility. Several coordination strategies, both with the optimized logic and the current logic, are evaluated in simulation.
READ LESS

Summary

The Traffic Alert and Collision Avoidance System (TCAS) is mandated on all large transport aircraft to reduce mid-air collision risk. Since its introduction, no mid-air collisions between TCAS-equipped aircraft have occurred in the United States. However, General Aviation (GA) aircraft are generally not equipped with TCAS and experience collisions several...

READ MORE
Collision avoidance for general aviation

Enhanced regional situational awareness

June 1, 2007
  |
Journal Article
Author:
Curtis W. Davis III
…
Published in:
Lincoln Laboratory Journal, Vol. 16, No. 2, June 2007, pp. 355-380.
Topic:
aircraft
R&D area:
R&D group:

Summary

Airspace protection in the capital area is provided by an Integrated Air Defense System (IADS) created through the coordinated response of U.S. government and local law-enforcement agencies, including the Department of Defense, the Department of Homeland Security, the Federal Aviation Administration, and the Capitol Police. The IADS includes U.S. Coast Guard helicopters, fighter aircraft, and airborne early-warning aircraft cued by surveillance radars. Under Operation Noble Eagle, the response to a threat includes warning flares deployed from fighter aircraft and, ultimately, the use of surface and air-launched missiles. Selecting the appropriate response requires a means for rapidly assessing the aircraft threat. New and existing sensors must be simultaneously cued to the target of interest and integrated with existing sources of information to display a common-air-picture display to support the decision makers. This article describes the development of an Enhanced Regional Situation Awareness system, an integrated sensing and decision support system developed for the complex and busy airspace surrounding the National Capital Region.
READ LESS

Summary

Airspace protection in the capital area is provided by an Integrated Air Defense System (IADS) created through the coordinated response of U.S. government and local law-enforcement agencies, including the Department of Defense, the Department of Homeland Security, the Federal Aviation Administration, and the Capitol Police. The IADS includes U.S. Coast...

READ MORE
Enhanced regional situational awareness

The development of phased-array radar technology

January 1, 2000
  |
Journal Article
Author:
Alan J. Fenn
…
Published in:
Lincoln Laboratory Journal, Vol. 12, No. 2, 2000, pp. 321-340.
Topic:
integrated circuit
R&D area:
R&D group:

Summary

Lincoln Laboratory has been involved in the development of phased-array radar technology since the late 1950s. Radar research activities have included theoretical analysis, application studies, hardware design, device fabrication, and system testing. Early phased-array research was centered on improving the national capability in phased-array radars. The Laboratory has developed several test-bed phased arrays, which have been used to demonstrate and evaluate components, beamforming techniques, calibration, and testing methodologies. The Laboratory has also contributed significantly in the area of phased-array antenna radiating elements, phase-shifter technology, solid-state transmit-and-receive modules, and monolithic microwave integrated circuit (MMIC) technology. A number of developmental phased-array radar systems have resulted from this research, as discussed in other articles in this issue. A wide variety of processing techniques and system components have also been developed. This article provides an overview of more than forty years of this phased-array radar research activity.
READ LESS

Summary

Lincoln Laboratory has been involved in the development of phased-array radar technology since the late 1950s. Radar research activities have included theoretical analysis, application studies, hardware design, device fabrication, and system testing. Early phased-array research was centered on improving the national capability in phased-array radars. The Laboratory has developed several...

READ MORE
The development of phased-array radar technology

Evaluation of the MTD in a high-clutter environment

January 1, 1980
  |
Conference Paper
Author:
John R. Anderson
…
Published in:
IEEE 1980 Int. Radar Conf., 28-30 April 1980, Arlington, VA, pp. 219-224.
Topic:
radar
R&D area:
R&D group:

Summary

The MTD (Moving Target Detector) is an automated radar signal and data processing system designed to improve the performance of air surveillance radars in various forms of clutter while providing a low output false alarm rate. This paper briefly describes the architecture of the MTD processor and presents the results of a field evaluation of the system using the ASR-7 terminal radar at Burlington, Vermont.
READ LESS

Summary

The MTD (Moving Target Detector) is an automated radar signal and data processing system designed to improve the performance of air surveillance radars in various forms of clutter while providing a low output false alarm rate. This paper briefly describes the architecture of the MTD processor and presents the results...

READ MORE
Evaluation of the MTD in a high-clutter environment
1-8 of 8