Search

Contextual awareness refers to having autonomous robots reason about their capabilities and limitation and improve on those limitations through the help of others. This project explored three areas of contextual awareness detecting when anomalous behaviors occur, reacting to situations where plans are failing, and learning new plans through human demonstration. Each of these areas is important in achieving robust, reliable robot autonomy. The work on detecting anomalous behavior focused on finding subtle anomalies that could not be detected from single events; the work on reacting to failing plans focused on deciding when to switch between risk-neutral and risk-seeking policies, for domains in which the goal is to achieve above a certain threshold of reward; and the work on learning new plans focused on complex manipulator trajectories, where multiple human examples are combined so as to smooth out noise in the examples without losing important details. The first and third areas were demonstrated using actual robots; the second area was demonstrated using a video game simulator.

Robots are just starting to appear in peopled environments but in order to be accepted by humans, they should obey basic people’s social rules. In particular, they have to be able to move around without disturbing people. This means that they have to obey the social rules that manage the movement of people, for example following virtual lanes when moving through corridors, not crossing in front of moving people, etc. In this paper some of these aspects are explained, as well as the implementation of preliminaries works to implement the proposed solutions are described. So, a slight modification to the Lane-Curvature Method is presented to improve the behavior of a mobile robot when crossing people in a corridor. Other works needed to test this modifications in the robot Amelia of the Reliable Autonomous Systems Lab, as the implementation of a fuzzy controller, are also described in this paper

In this paper, we describe an approach to scheduling under uncertainty that achieves scalability through a coupling of deterministic and probabilistic reasoning. Our specific focus is a class of oversubscribed scheduling problems where the goal is to maximize the reward earned by a team of agents in a distributed execution environment. There is uncertainty in both the duration and outcomes of executed activities. To ensure scalability, our solution approach takes as its starting point an initial deterministic schedule for the agents, computed using expected duration reasoning. This initial agent schedule is probabilistically analyzed to find likely points of failure, and then selectively strengthened based on this analysis. For each scheduled activity, the probability of failing and the impact that failure would have on the schedule's overall reward are calculated and used to focus schedule strengthening actions. Such actions generally entail fundamental trade-offs; for example, modifications that increase the certainty that a high-reward activity succeeds may decrease the schedule slack available to accommodate uncertainty during execution. We describe a principled approach to handling these trade-offs based on the schedule's "expected reward," using it as a metric to ensure that all schedule modifications are ultimately beneficial. Finally, we present experimental results obtained using a multi-agent simulation environment, which confirm that executing schedules strengthened in this way result in significantly higher rewards than are achieved by executing the corresponding initial schedules., Presented at the International Joint Conference on Artificial Intelligence (IJCAI-09) (21st), held in Pasadena, CA, on 11-17 Jul 2009. U.S. Government or Federal Rights License

To fulfill the needs of its deep space exploration program, NASA is actively supporting research and development in autonomy software. However, the reliable and cost-effective development and validation of autonomy systems poses a tough challenge. Traditional scenario-based testing methods fall short because of the combinatorial explosion of possible situations to be analyzed, and formal verification techniques typically require a tedious, manual modeling by formal method experts. This paper presents the application of formal verification techniques in the development of autonomous controllers based on Livingstone, a model-based health-monitoring system that can detect and diagnose anomalies and suggest possible recovery actions. We present a translator that converts the models used by Livingstone into specifications that can be verified with the SMV model checker. The translation frees the Livingstone developer from the tedious conversion of his design to SMV, and isolates him from the technical details of the SMV program. We describe different aspects of the translation and briefly discuss its application to several NASA domains.

In an attempt to solve as much of the AAAI Robot Challenge as possible, five research institutions representing academia, industry and government, integrated their research on a pair of robots named GRACE and GEORGE. This paper describes the second year effort by the GRACE team, the various techniques each participant brought to GRACE, and the integration effort itself., Prepared in collaboration with Carnegie Mellon University; Northwestern University; Metrica, Inc.; and Swarthmore College. The original document contains color images.

This report describes a paradigm for combining associational and causal reasoning to achieve efficient and robust problem-solving behavior. The Generate, Test and Debug (GTD) paradigm generates initial hypotheses using associational (heuristic) rules. The tester verifies hypotheses, supplying the debugger with causal explanations for bugs found if the test fails. The debugger uses domain-independent causal reasoning techniques to repair hypotheses, analyzing domain models and the causal explanations produced by the tester to determine how to replace faulty assumptions made by the generator. We analyze the strengths and weaknesses of associational and causal reasoning techniques, and present a theory of debugging plans and interpretations. The GTD paradigm has been implemented and tested in the domains of geologic interpretation, the blocks world, and Tower of Hanoi problems.

Geologic interpretation is the task of inferring a sequence of events to explain how a given geologic region could have been formed. This report describes the design and implementation of one part of a geologic interpretation problem solver -- a system which uses a simulation technique called imagining to check the validity of a candidate sequence of events. Imagining uses a combination of qualitative and quantitative simulations to reason about the changes which occured to the geologic region. The spatial changes which occur are simulated by constructing a sequence of diagrams. The quantitative simulation needs numeric parameters which are determined by using the qualitative simulation to establish the cumulative changes to an object and by using a description of the current geologic region to make quantitative measurements. The diversity of reasoning skills used in imagining has necessitated the development of multiple representations, each specialized for a different task. Representations to facilitate doing temporal, spatial and numeric reasoning are described in detail. We have also found it useful to explicitly represent processes. Both the qualitative and quantitative simulations use a discrete 'layer cake' model of geologic processes, but each uses a separate representation, specialized to support the type of simulation. These multiple representations have enabled us to develop a powerful, yet modular, system for reasoning about change.

This paper explores representations used to reason about objects which change over time and the processes which cause changes. Specifically, we are interested in solving a problem known as geologic interpretation. To help solve this problem, we have developed a simulation technique, which we call imagining. Imagining takes a sequence of events and simulates them by drawing diagrams. In order to do this imagining, we have developed two representations of objects, one involving histories and the other involving diagrams, and two corresponding representations of physical processes, each suited to reasoning about one of the object representations. These representations facilitate both spatial and temporal reasoning.

The AAAI Robot Challenge was established four years ago as a "grand challenge" for mobile robots. The main objectives of the Challenge are to (a) provide a task that will demonstrate a high level of intelligence and autonomy for robots acting in a natural, peopled, dynamic environment; (b) stimulate state-of-the-art robotics research to address this task; and (c) use robot demonstrations to educate the public about the exciting and difficult challenges of robotics research. When the Challenge was designed, it was anticipated that no single research institution would have adequate resources to meet the Challenge on its own. The Challenge task is to find the registration booth and register at the National Conference on Artificial Intelligence, interact with other attendees, and give a technical talk on itself in an assigned room, and at an assigned time. Ideally, the robot should be given no more information than any other participant arriving in a new city to attend a major technical conference. In particular, that means that the robot should not know the layout of the convention center beforehand, and the environment should not be modified. Practically, however, the organizers understand that compromises and flexibility will be necessary in order to get current state-of-the-art robots to achieve such a task., Prepared in collaboration with Naval Research Laboratory. Prepared in cooperation with Northwestern University, Metrica, Inc., and Swarthmore College. The original document contains color images.

In a very original study on human-robot interaction, researchers at Carnegie Mellon University conducted a series of experiments to determine if people would be more willing to interact with a robot with human-like features as opposed to one without. The particular "behaviors tested were the ability to convey expression with a humanoid face and the ability to indicate attention by turning towards the person that the robot is addressing." This publication discusses the specifics of these experiments and the results. The basis of the study involved a robot asking a passerby if they would like to participate in a poll. Interestingly, both having a face and having tracking behavior effected how many people stopped to talk to the robot.

To be truly useful, mobile robots need to be fairly autonomous and easy to control. This is especially true in situations where multiple robots are used, due to the increase in sensory information and the fact that the robots can interfere with one another. This paper describes a system that integrates autonomous navigation, a task executive, task planning, and an intuitive graphical user interface to control multiple, heterogeneous robots. We have demonstrated a prototype system that plans and coordinates the deployment of teams of robots. Testing has shown the effectiveness and robustness of the system, and of the coordination strategies in particular., Sponsored in part by DARPA. Presented at the International Conference on Intelligent Robots and Systems (IROS) held in Takamatsu, Japan in October 2000. Published in the Proceedings of the International Conference on Intelligent Robots and Systems (IROS), 2000.