Matthew J. Bays

A solution to the multiple aspect coverage problem

We introduce a novel task that arises in underwater search and inspection. In this task, an underwater vehicle must re-acquire and identify clusters of discrete objects. The challenge is to generate an efficient path for the vehicle given a probabilistic description of potential target locations. We propose an algorithm that generates an efficient path and show that the algorithm is superior to standard approaches.

Automated scheduling and mission visualization for mine countermeasure operations

We present a novel schedule optimization framework to plan tasks for multiple unmanned surface vessels (USVs) and unmanned underwater vehicles (UUVs) performing a mine countermeasure (MCM) mission. The framework employs mixed-integer linear programming (MILP) to determine efficient schedules. In the framework, the USVs are capable of transporting the UUVs, processing contact information, and neutralizing targets. The UUVs perform survey as well as reacquire and identify (RI) operations. Additionally, we present a novel MCM mission visualizer. The goal of the visualizer is to display in 3D an entire planned MCM scenario for use in scenario evaluation as well as provide a detailed multi-vehicle testing platform for MCM-related vehicle and autonomy research and development.

Stochastic target interception in non-convex domain using MILP

In this paper we present a planning approach for the stochastic target interception problem, in which, a team of mobile sensor agents is tasked with intercepting multiple targets. We extend our previous work on stochastic target interception to non-convex domains and propose a cost that addresses minimum time requirement for probabilistically intercepting all the targets if possible over a finite horizon. Indeed, our optimization problem for the stochastic case has similar computational costs as the optimization program for the corresponding deterministic case. Our solution presumes that the system can be approximated by linear dynamics and Gaussian noise, with Gaussian localization uncertainty.

Service agenttransport agent task planning incorporating robust scheduling techniques

As the use of cooperative, heterogeneous teams of autonomous robots to perform tasks such as autonomous package delivery and long-duration ocean sampling becomes more prevalent, the is a quickly-emerging need to study the high-level interaction of specialized robotic agents that perform service tasks, and specialized transport agents that transport the service agents between service areas. If the service routes, docking, and deployment schedules are not carefully planned, the overall schedule is inefficient at best, and possibly even infeasible due to fuel limitations at worst. We introduce a new problem in the area of scheduling and route planning operations called the service agent transport problem (SATP). Within the SATP, autonomous service agents must perform tasks at a number of locations. The agents are free to move between locations, however, the agents may also be transported throughout the region by a limited number of faster-moving transport agents. The goal of the SATP is to plan a schedule of service agent and transport agent actions such that all locations are serviced in the shortest amount of time. We believe the SATP formulation is unique because there is strong coupling between vehicle constraints as well as between the task allocation component of the problem and the scheduling component of the problem. We present a solution to the problem using a mixed-integer linear programming optimization framework and compare several complexity reduction heuristics to full optimization. Additionally, we include methods to account for relative uncertainty in the duration of planned tasks in such a manner as to balance the risk of schedule slips (conflicts) to the risk of creating an overly conservative and sub-optimal schedule. We introduce a new problem in multi-agent task allocation and scheduling.Problem involves multiple service agents and transport agents.We model problem in a mixed-integer linear programming framework.We extend problem to a robust form leveraging Bayes risk.

Maritime autonomy in a box: Building a quickly-deployable autonomy solution using the Docker container environment

We present a method to dramatically reduce the level of effort and lead time required to take autonomy algorithms from initial development to field experiments when using shared assets. The method leverages the Docker containerization environment coupled with automatic configuration and deployment modules and a standardized autonomy framework. The result is a quickly-deployable, easily reconfigurable, and vehicle-agnostic autonomy solution for use when assets are shared and repeatedly re-baselining the system is necessary. For an initial implementation of the containerization system, we leverage the Mission-Oriented Operating Suite - Interval Programming autonomy framework.

A solution to the service agent transport problem

We introduce a new problem in the area of scheduling and route planning operations called the service agent transport problem (SATP). Within the SATP, autonomous service agents must perform tasks at a number of locations. The agents are free to move between locations, however, the agents may also be transported throughout the region by a limited number of faster-moving transport agents. The goal of the SATP is to plan a schedule of service agent and transport agent actions such that all locations are serviced in the shortest amount of time. We pose the problem using a mixed-integer linear programming optimization framework, and compare several complexity reduction heuristics to the full optimization.

Toward a mission value for subsea search with bottom-type variability

We propose a value function that can be used to evaluate candidate search paths for mobile sensor agents that seek to find specific objects. Our work is motivated by subsea applications where the mobile sensor agent surveys the seafloor using sonar. We presume that typical sensor performance characteristics are known, including probability of detection and probability of false alarm. Since variations in the environment can affect sensor performance, we also address the case that a stochastic description of the environment is available.

An approach to multi-agent area protection using bayes risk

We introduce a novel approach to controlling the motion of a team of agents so that they jointly minimize a cost function utilizing Bayes risk. We use a particle-based approach and approximations that allow us to express the optimization problem as a mixed-integer linear program. We illustrate this approach with an area protection problem in which a team of mobile agents must intercept mobile targets before the targets enter a specified area. Bayes risk is a useful measure of performance for applications where agents must perform a classification task. By minimizing Bayes risk, agents are able to explicitly account for the cost of incorrect classification. In our application, a team of mobile agents must classify potential mobile targets as threat or safe based on the likelihood the targets will enter the specified area. The agents must also maneuver to intercept targets that are classified as threat.

Multiple agent coordination for stochastic target interception using MILP

In this paper we present an approach to solving a stochastic multi-target interception problem. In the multi-target interception problem, a team of mobile sensors is tasked with intercepting a set of potential targets to reduce appropriately assigned damage cost. Our principal contribution is to express a stochastic version of the problem with a generalized cost as a mixed-integer linear program so that optimal sensor motion can be computed efficiently. Indeed, our optimization program for the stochastic problem has similar computational costs as the optimization program for the corresponding deterministic problem. Our solution presumes that the system can be approximated by linear dynamics and Gaussian noise, with Gaussian localization uncertainty.

Partially-Decoupled Service Agent - Transport Agent Task Allocation and Scheduling

We present an approach to performing efficient schedule generation of a heterogeneous team of autonomous agents in a service agent - transport agent scenario where the task allocation and scheduling components are partially-decoupled. In the scenario, service agents must perform tasks at a number of locations. The agents are free to move between locations, however the agents may also be transported throughout the region by a limited number of faster-moving transport agents. The goal of the problem is to plan a schedule of service agent and transport agent actions such that all locations are serviced in the shortest amount of time. While in previous work we formulated the problem as a holistic mixed-integer linear program, we present a novel method to solve the problem in a hierarchical and partially-decoupled manner for faster optimization and to require less information to be processed and communicated in a centralized manner to perform the schedule planning. The original solution method required up to 20 minutes to obtain an efficient solution. The new methodology, using hierarchical task allocation and a bidding-based scheduling algorithm, can create an efficient solution in seconds.