Ethan Stump

Persistent surveillance with a team of MAVs

In this paper, we focus on the detailing of a system architecture capable of addressing the problem of persistent surveillance with a team of autonomous micro-aerial vehicles (MAVs). We detail the problem of interest, discuss system requirements, and provide an overview of our approach. The remainder of the paper is dedicated to the system design and evaluation on a team of quadrotors in simulation and experiments.

Multi-robot persistent surveillance planning as a Vehicle Routing Problem

We consider a persistent surveillance problem as one of finding sequences of visits to discrete sites in a periodic fashion, cast it as a Vehicle Routing Problem with Time Windows, and solve it using exact methods developed in the operations research community. The work is a successful application of recent advances in combinatorial optimization for logistics problems but in the context of a robotics application taking place in continuous time. We apply these methods to the task of surveying a building using multiple UAVs and perform a long-term simulation developed to mimic a hardware testbed currently under development.

Visibility-based deployment of robot formations for communication maintenance

We consider the problem of deploying robots in formations that ensure network connectivity between a fixed base station and a set of independent agents wandering in the environment. We adopt a communications model that requires line-of-sight and then solve for robot placements by finding mutually-visible configurations in a polygonal decomposition of the environment map. Both the static deployment case and the case of finding deployments that minimize total robot movement are considered. We provide algorithms for the moving agent case, consider their performance on various discretizations for a range of problem sizes, and discuss our experimental implementation of the presented ideas.

D4L: Decentralized dynamic discriminative dictionary learning

We consider discriminative dictionary learning in a distributed online setting, where a network of agents aims to learn, from sequential observations, statistical model parameters jointly with data-driven signal representations. We formulate this problem as a distributed stochastic program with a nonconvex objective that quantifies the merit of the choice of model parameters and dictionary. We consider the use of a block variant of the Arrow–Hurwicz saddle point algorithm to solve this problem, which exploits factorization properties of the Lagrangian to yield a protocol in that only requires exchange of model information among neighboring nodes. We show that decisions made with this saddle point algorithm asymptotically achieve a first-order stationarity condition on average. The learning rate depends on the signal source, network structure, and discriminative task. We illustrate the algorithm performance for solving a large-scale image classification task on a network of interconnected servers and observe that practical performance is comparable to a centralized approach. We further apply this method to the problem of a robotic team seeking to autonomously navigate in an unknown environment by predicting unexpected maneuvers, demonstrating the proposed algorithms utility in a field setting.

Application of Multi-Robot Systems to Disaster-Relief Scenarios with Limited Communication

In this systems description paper, we present a multi-robot solution for intelligence-gathering tasks in disaster-relief scenarios where communication quality is uncertain. First, we propose a formal problem statement in the context of operations research. The hardware configuration of two heterogeneous robotic platforms capable of performing experiments in a relevant field environment and a suite of autonomy-enabled behaviors that support operation in a communication-limited setting are described. We also highlight a custom user interface designed specifically for task allocation amongst a group of robots towards completing a central mission. Finally, we provide an experimental design and extensive, preliminary results for studying the effectiveness of our system.

Mapping with a ground robot in GPS denied and degraded environments

A robot system operating in an unknown environment must be able to track its position to perform its mission. Vehicles with a consistent view of the sky, e.g., aerial or water surface platforms, can reliably make use of GPS signals to correct accumulated error from inertial measurements and feature-based mapping techniques. However, ground robots that must operate across a wide range of environments suffer from additional constraints which degrade the performance of GPS such as multipath and occlusion. In this paper, we present a methodology for incorporating GPS measurements into a feature-based mapping system for two purposes: providing geo-referenced coordinates for high-level mission execution and correcting accumulated map error over long-term operation. We will present both the underlying system and experimental results from a variety of relevant environments such as military training facilities and large-scale mixed indoor and outdoor environments.

Online learning for characterizing unknown environments in ground robotic vehicle models

In pursuit of increasing the operational tempo of a ground robotics platform in unknown domains, we consider the problem of predicting the distribution of structural state-estimation error due to poorly-modeled platform dynamics as well as environmental effects. Such predictions are a critical component of any modern control approach that utilizes uncertainty information to provide robustness in control design. We use an online learning algorithm based on matrix factorization techniques to fit a statistical model of error that provides enough expressive power to enable prediction directly from motion control signals and low-level visual features. Moreover, we empirically demonstrate that this technique compares favorably to predictors that do not incorporate this information.

The privileged sensing framework: A principled approach to improved human-autonomy integration

ABSTRACT A primary goal for human-autonomy integration (HAI) is to balance the strengths of human and autonomy in order to achieve performance objectives more efficiently and robustly than either the human or autonomous agents would independently. This paper proposes the Privileged Sensing Framework (PSF) as a novel approach to HAI. This approach is based on the concept of dynamically ‘privileging’ information during the process of integration by dynamically bestowing special rights based on the characteristics of each individual agent, the task context, and the performance goals. The proposed framework is tested through a series of simulation experiments that provide a clear demonstration of increased accuracy and throughput of human-autonomy performance. These proof-of-concept simulations provide initial evidence of the utility of the PSF. Continued development of this approach has the potential to revolutionise capabilities of multi-agent cooperative teams across a broad range of applications.

Experimental analysis of models for trajectory generation on tracked vehicles

We begin to bridge the gap between high-level motion planning and execution by adopting models to abstract the complicated skid-steer vehicle dynamics and evaluating their suitability as motion predictors for a feed-forward control framework. We consider three kinematic motion models and a drivetrain model in experiments on two surface types with a small tracked vehicle. We perform statistical analysis of the predictive accuracy of these models when used to create optimal open-loop plans for a set of canonical maneuvers and discuss the applicability of these models for a closed-loop control framework.

Task-driven dictionary learning in distributed online settings

We consider task-driven dictionary learning in a decentralized dynamic setting. Here a network of agents while sequentially receiving local information aims to learn a common data-driven signal representation and model parameters. We formulate this problem as a distributed stochastic program with a non-convex objective and present a block variant of the Arrow-Hurwicz saddle point algorithm to solve it. Using Lagrange multipliers to penalize the discrepancy between them, only neighboring nodes exchange model information. We show that decisions made with this saddle point algorithm asymptotically converge to a stationarity condition in expectation under certain conditions. The learning rate depends on the signal source, network, and discriminative task. We illustrate the algorithm performance in an online multi-agent setting for a collaborative image classification task, demonstrating that the performance is comparable to the centralized case and depends on the network topology over which it is run.