Engineering topics

Washington, Richard

Publications and source records attributed to Washington, Richard.

14 records

NASA NTRS2018

A Detailed Characterization of the Saharan Dust Collected During the Fennec Campaign in 2011: In Situ Ground-Based and Laboratory Measurements

Millions of tons of mineral dust are lifted by the wind from arid surfaces and transported around the globe every year. The physical and chemical properties of the mineral dust are needed to better constrain remote sensing observations and are of fundamental importance for the understanding of dust atmospheric processes. Ground-based in situ measurements and in situ filter collection of Saharan dust were obtained during the Fennec campaign in the central Sahara in 2011. This paper presents results of the absorption and scattering coefficients, and hence single scattering albedo (SSA), of the Saharan dust measured in real time during the last period of the campaign and subsequent laboratory analysis of the dust samples collected in two supersites, SS1 and SS2, in Algeria and in Mauritania, respectively. The samples were taken to the laboratory, where their size and aspect ratio distributions, mean chemical composition, spectral mass absorption efficiency, and spectral imaginary refractive index were obtained from the ultraviolet (UV) to the near-infrared (NIR) wavelengths. At SS1 in Algeria, the time series of the scattering coefficients during the period of the campaign show dust events exceeding 3500 Mm (exp -1), and a relatively high mean SSA of 0.995 at 670 nm was observed at this site. The laboratory results show for the fine particle size distributions (particles diameter <5μm and mode diameter at2–3 μm) in both sites a spectral dependence of the imaginary part of the refractive index Im(m) with a bow-like shape, with increased absorption in UV as well as in the shortwave infrared. The same signature was not observed, however, in the mixed particle size distribution (particle diameter<10 μm and mode diameter at 4 μm) in Algeria. Im(m) was found to range from 0.011 to 0.001i for dust collected in Algeria and 0.008 to 0.002i for dust collected in Mauritania over the wavelength range of 350-2500 nm. Differences in the mean elemental composition of the dust collected in the supersites in Algeria and in Mauritania and between fine and mixed particle size distributions were observed from EDXRF measurements, although those differences cannot be used to explain the optical properties variability between the samples. Finally, particles with low-density typically larger than 10 μm in diameter were found in some of the samples collected at the supersite in Mauritania, but these low-density particles were not observed in Algeria.

Saharan dust measured in real time↗

NASA NTRS2006

Architecture for Control of the K9 Rover

Software featuring a multilevel architecture is used to control the hardware on the K9 Rover, which is a mobile robot used in research on robots for scientific exploration and autonomous operation in general. The software consists of five types of modules: Device Drivers - These modules, at the lowest level of the architecture, directly control motors, cameras, data buses, and other hardware devices. Resource Managers - Each of these modules controls several device drivers. Resource managers can be commanded by either a remote operator or the pilot or conditional-executive modules described below. Behaviors and Data Processors - These modules perform computations for such functions as planning paths, avoiding obstacles, visual tracking, and stereoscopy. These modules can be commanded only by the pilot. Pilot - The pilot receives a possibly complex command from the remote operator or the conditional executive, then decomposes the command into (1) more-specific commands to the resource managers and (2) requests for information from the behaviors and data processors. Conditional Executive - This highest-level module interprets a command plan sent by the remote operator, determines whether resources required for execution of the plan are available, monitors execution, and, if necessary, selects an alternate branch of the plan.

Bresina, John L.↗

NASA NTRS2006

The K9 On-Board Rover Architecture

This paper describes the software architecture of NASA Ames Research Center s K9 rover. The goal of the onboard software architecture team was to develop a modular, flexible framework that would allow both high- and low-level control of the K9 hardware. Examples of low-level control are the simple drive or pan/tilt commands which are handled by the resource managers, and examples of high-level control are the command sequences which are handled by the conditional executive. In between these two control levels are complex behavioral commands which are handled by the pilot, such as drive to goal with obstacle avoidance or visually servo to a target. This paper presents the design of the architecture as of Fall 2000. We describe the state of the architecture implementation as well as its current evolution. An early version of the architecture was used for K9 operations during a dual-rover field experiment conducted by NASA Ames Research Center (ARC) and the Jet Propulsion Laboratory (JPL) from May 14 to May 16, 2000.

Bresina, John L.↗

NASA NTRS2004

Dynamic Programming for Structured Continuous Markov Decision Problems

We describe an approach for exploiting structure in Markov Decision Processes with continuous state variables. At each step of the dynamic programming, the state space is dynamically partitioned into regions where the value function is the same throughout the region. We first describe the algorithm for piecewise constant representations. We then extend it to piecewise linear representations, using techniques from POMDPs to represent and reason about linear surfaces efficiently. We show that for complex, structured problems, our approach exploits the natural structure so that optimal solutions can be computed efficiently.

Dearden, Richard↗

NASA NTRS2003

Decision-Theoretic Control of Planetary Rovers

Planetary rovers are small unmanned vehicles equipped with cameras and a variety of sensors used for scientific experiments. They must operate under tight constraints over such resources as operation time, power, storage capacity, and communication bandwidth. Moreover, the limited computational resources of the rover limit the complexity of on-line planning and scheduling. We describe two decision-theoretic approaches to maximize the productivity of planetary rovers: one based on adaptive planning and the other on hierarchical reinforcement learning. Both approaches map the problem into a Markov decision problem and attempt to solve a large part of the problem off-line, exploiting the structure of the plan and independence between plan components. We examine the advantages and limitations of these techniques and their scalability.

Zilberstein, Shlomo↗

NASA NTRS2001

VIPER: Virtual Intelligent Planetary Exploration Rover

Simulation and visualization of rover behavior are critical capabilities for scientists and rover operators to construct, test, and validate plans for commanding a remote rover. The VIPER system links these capabilities. using a high-fidelity virtual-reality (VR) environment. a kinematically accurate simulator, and a flexible plan executive to allow users to simulate and visualize possible execution outcomes of a plan under development. This work is part of a larger vision of a science-centered rover control environment, where a scientist may inspect and explore the environment via VR tools, specify science goals, and visualize the expected and actual behavior of the remote rover. The VIPER system is constructed from three generic systems, linked together via a minimal amount of customization into the integrated system. The complete system points out the power of combining plan execution, simulation, and visualization for envisioning rover behavior; it also demonstrates the utility of developing generic technologies. which can be combined in novel and useful ways.

Edwards, Laurence↗

NASA NTRS2000

Robustness via Run-Time Adaptation of Contingent Plans

In this paper, we discuss our approach to making the behavior of planetary rovers more robust for the purpose of increased productivity. Due to the inherent uncertainty in rover exploration, the traditional approach to rover control is conservative, limiting the autonomous operation of the rover and sacrificing performance for safety. Our objective is to increase the science productivity possible within a single uplink by allowing the rover's behavior to be specified with flexible, contingent plans and by employing dynamic plan adaptation during execution. We have deployed a system exhibiting flexible, contingent execution; this paper concentrates on our ongoing efforts on plan adaptation, Plans can be revised in two ways: plan steps may be deleted, with execution continuing with the plan suffix; and the current plan may be merged with an "alternate plan" from an on-board library. The plan revision action is chosen to maximize the expected utility of the plan. Plan merging and action deletion constitute a more conservative general-purpose planning system; in return, our approach is more efficient and more easily verified, two important criteria for deployed rovers.

Bresina, John L.↗

NASA NTRS2000

On-Board Real-Time State and Fault Identification for Rovers

For extended autonomous operation, rovers must identify potential faults to determine whether its execution needs to be halted or not. At the same time, rovers present particular challenges for state estimation techniques: they are subject to environmental influences that affect senior readings during normal and anomalous operation, and the sensors fluctuate rapidly both because of noise and because of the dynamics of the rover's interaction with its environment. This paper presents MAKSI, an on-board method for state estimation and fault diagnosis that is particularly appropriate for rovers. The method is based on a combination of continuous state estimation, wing Kalman filters, and discrete state estimation, wing a Markov-model representation.

Washington, Richard↗

NASA NTRS2000

Expected Utility Distributions for Flexible, Contingent Execution

This paper presents a method for using expected utility distributions in the execution of flexible, contingent plans. A utility distribution maps the possible start times of an action to the expected utility of the plan suffix starting with that action. The contingent plan encodes a tree of possible courses of action and includes flexible temporal constraints and resource constraints. When execution reaches a branch point, the eligible option with the highest expected utility at that point in time is selected. The utility distributions make this selection sensitive to the runtime context, yet still efficient. Our approach uses predictions of action duration uncertainty as well as expectations of resource usage and availability to determine when an action can execute and with what probability. Execution windows and probabilities inevitably change as execution proceeds, but such changes do not invalidate the cached utility distributions, thus, dynamic updating of utility information is minimized.

Bresina, John L.↗

NASA NTRS1999

State Identification for Planetary Rovers: Learning and Recognition

A planetary rover must be able to identify states where it should stop or change its plan. With limited and infrequent communication from ground, the rover must recognize states accurately. However, the sensor data is inherently noisy, so identifying the temporal patterns of data that correspond to interesting or important states becomes a complex problem. In this paper, we present an approach to state identification using second-order Hidden Markov Models. Models are trained automatically on a set of labeled training data; the rover uses those models to identify its state from the observed data. The approach is demonstrated on data from a planetary rover platform.

Aycard, Olivier↗

NASA NTRS1999

Autonomous Rovers for Mars Exploration

Rovers will play a critical role in the exploration of Mars. Near-term mission plans call for long traverses over unknown terrain, robust navigation and instrument placement, and reliable operations for extended periods of time. Longer-term missions may visit multiple science sites in a single day and perform opportunistic science data collection, as well as complex scouting, construction, and maintenance tasks in preparation for an eventual human presence. The Pathfinder mission demonstrated the potential for robotic Mars exploration but at the same time indicated the need for more rover autonomy. The highly ground-intensive control with infrequent communication and high latency limited the effectiveness of the Sojourner rover. When failures occurred, Sojourner often sat idle for extended periods of time, awaiting further commands from earth. In future missions, the tasks will be more complex and extended; hence there will be even more situations where things do not go exactly as planned. Significant advances in rover autonomy are needed to cope with increasing task complexity and greater execution uncertainty. Towards this end, we have designed an on-board executive architecture that incorporates robust operation, resource utilization, and failure recovery. In addition, we have designed ground tools to produce and refine contingent schedules that take advantage of the on-board architecture's flexible execution characteristics. Together, the on-board executive and the ground tools constitute an integrated rover autonomy architecture. This work draws from our experience with the Deep Space One autonomy experiment, with enhancements to ensure robust operation in the face of the unpredictable, complex environment that the rover will encounter on Mars. The rover autonomy architecture is currently being developed and deployed on the Marsokhod rover platform at NASA Ames Research Center. The capabilities of the rover autonomy architecture to support autonomous operations will be demonstrated concretely in upcoming field tests.

Anderson, Corin↗

NASA NTRS1998

Markov Tracking for Agent Coordination

Partially observable Markov decision processes (POMDPs) axe an attractive representation for representing agent behavior, since they capture uncertainty in both the agent's state and its actions. However, finding an optimal policy for POMDPs in general is computationally difficult. In this paper we present Markov Tracking, a restricted problem of coordinating actions with an agent or process represented as a POMDP Because the actions coordinate with the agent rather than influence its behavior, the optimal solution to this problem can be computed locally and quickly. We also demonstrate the use of the technique on sequential POMDPs, which can be used to model a behavior that follows a linear, acyclic trajectory through a series of states. By imposing a "windowing" restriction that restricts the number of possible alternatives considered at any moment to a fixed size, a coordinating action can be calculated in constant time, making this amenable to coordination with complex agents.

Washington, Richard↗

NASA NTRS1998

Making the Impossible Possible: Strategies for Fast POMDP Monitoring

Systems modeled as partially observable Markov decision processes (POMDPs) can be tracked quickly with three restrictions: all actions are grouped together, the out-degree of each system state is bounded by a constant, and the number of non-zero elements in the belief state is bounded by a (different) constant. With these restrictions, the tracking algorithm operates in constant time and linear space. The first restriction assumes that the action itself is unobservable. The second restriction defines a subclass of POMDPs that covers however a wide range of problems. The third restriction is an approximation technique that can lead to a potentially vexing problem: an observation may be received that has probability according to the restricted belief state. This problem of impossibility will cause the belief state to collapse. In this paper we discuss the tradeoffs between the constant bound on the belief state and the quality of the solution. We concentrate on strategies for overcoming the impossibility problem and demonstrate initial experimental results that indicate promising directions.

Washington, Richard↗

NASA NTRS1994

Abstraction Planning in Real Time

When a planning agent works in a complex, real-world domain, it is unable to plan for and store all possible contingencies and problem situations ahead of time. The agent needs to be able to fall back on an ability to construct plans at run time under time constraints. This thesis presents a method for planning at run time that incrementally builds up plans at multiple levels of abstraction. The plans are continually updated by information from the world, allowing the planner to adjust its plan to a changing world during the planning process. All the information is represented over intervals of time, allowing the planner to reason about durations, deadlines, and delays within its plan. In addition to the method, the thesis presents a formal model of the planning process and uses the model to investigate planning strategies. The method has been implemented, and experiments have been run to validate the overall approach and the theoretical model.

ARTIFICIAL INTELLIGENCE↗