Person:

Goel, Ashok K.

Permanent Link

https://hdl.handle.net/1853/71296

Associated Organization(s)

Organizational Unit

School of Computational Science and Engineering

Full item page

Publication Search Results

Now showing 1 - 9 of 9

Playing Detective: Using AI for Sensemaking in Investigative Analysis

(Georgia Institute of Technology, 2009) Goel, Ashok K. ; Adams, Summer ; Cutshaw, Neil ; Sugandh, Neha

The sensemaking task in investigative analysis generates models that connect entities and events in an input stream of data. We describe two knowledge systems for aiding sensemaking in investigative analysis. The Spade system uses crime schemas to generate an explanatory hypothesis and past cases to validate the hypothesis. The STAB system represents crime schemas as hierarchical scripts with goals and states. It generates multiple explanatory hypotheses for an input data stream containing interleaved sequences of events, recognizes intent in a specific event sequence, and calculates confidence values for the generated hypotheses. We view STAB and Spade as automated cognitive assistants to human analysts: they may support sensemaking in investigative analysis by generating and managing multiple competing hypotheses.
Learning About and Through Biologically Inspired Design

(Georgia Institute of Technology, 2008-06-22) Vattam, Swaroop ; Helms, Michael E. ; Goel, Ashok K. ; Yen, Jeannette ; Weissburg, Marc J.

Biologically inspired design (BID) uses biological systems as analogues to develop solutions for design problems. Although designers have been looking to nature for inspiration for eons, only recently is BID gaining in importance as a wide-spread movement in design for environmentally-conscious sustainable development (e.g., Benyus 1997). But it is the tendency of the “products” of BID to be radically innovative (Forbes 2005; French 1998; Vogel 2000) that makes BID an interesting case for research in design creativity.
Biologically-Inspired Innovation in Engineering Design: a Cognitive Study

(Georgia Institute of Technology, 2007) Vattam, Swaroop ; Helms, Michael E. ; Goel, Ashok K.

Biologically-inspired design uses analogous biological phenomena to develop solutions for engineering problems. Understanding, learning and practicing this approach to design is challenging because biologists and engineers speak different languages, have different perspectives on design, with different constraints on design problems and different resources for realizing an abstract design. In Fall 2006, we attended ME/ISyE/MSE/PTFe/BIOL 4803: Biologically-Inspired Design, an interdisciplinary introductory course for juniors and seniors offered at Georgia Tech. We collected course materials, took class notes, observed teacher-student and student-student interactions in the classroom. We also observed some sessions of a few interdisciplinary teams of students engaged in their design projects outside the classroom. We then analyzed the observations in terms of existing cognitive theories of design, modeling, and analogy. The goals of this cognitive study were to (1) understand the cognitive basis of biologically-inspired innovation in engineering design, (2) identify opportunities for enabling more effective learning of biologically-inspired design, and (3) examine the implications for developing computational tools for facilitating effective biologically-inspired design. This report summarizes our main observations about learning biologically-inspired design, and presents our preliminary analysis of biologically-inspired design in a classroom setting.
Using Spatial Structure in the Associative Retrieval of 2-D Line Drawings

(Georgia Institute of Technology, 2002) Yaner, Patrick W. ; Goel, Ashok K.

We consider the problem of associative image retrieval, focusing on retrieval of 2-D line drawings by example. We represent 2-D line drawings as semantic networks of spatial elements and relations among them. We describe a process for retrieving the drawings based on a structural analogy between the query and the stored images. We then present several methods of retrieving the drawings: the first family methods uses logical unification and resolution to accomplish the matching; the second family of methods heuristically prunes the stored drawings and then does the resolution and unification on the remaining drawings; two more methods treat the retrieval problem as a constraint satisfaction problem and use common CSP techniques for solving it; and the last two methods combine the heuristic step of the second method with the CSP technique of the third and fourth. We report on experimental results that compare the performance of these methods on computer-based libraries of drawings. A surprising result of our work is that for the fastest of these methods two stage retrieval appeared to offer no benefit over one stage retrieval.
Model-Based Reconfiguration of Schema-Based Reactive Control Architectures

(Georgia Institute of Technology, 1997) Chen, Zhong ; Goel, Ashok K. ; Rowland, Paul ; Stroulia, Eleni

Reactive methods of control get caught in local minima. Fortunately schema-based reactive control systems have built-in redundancy that enables multiple configurations with different modes. We describe a model-based method that exploits this redundancy, and, under certain conditions, reconfigures schema-based reactive control systems trapped in behavioral cycles due to the presence of local minima. The qualitative model specifies the functions and modes of the perceptual and motor schemas, and represents the reactive architecture as a structure-behavior-function model. The model-based method monitors the reactive processing, detects failures in the form of behavioral cycles, analyzes the processing trace, identifies potential modifications, and reconfigures the reactive architecture. We report on experiments with a simulated robot navigating a complex space.
Some Experimental Results in Multistrategy Navigation Planning

(Georgia Institute of Technology, 1995) Goel, Ashok K. ; Ali, Khaled Subhi ; Stroulia, Eleni

Spatial navigation is a classical problem in AI. In the paper, we examine three specific hypotheses regarding multistrategy navigation planning in visually engineered physical spaces containing discrete pathways: (1) For Hybrid robots capable of both deliberative planning and situated action, qualitative representations of topological knowledge are sufficient for enabling effective spatial navigation; (2) For deliberative planning, the case-based strategy of plan reuse generates plans more efficiently than the model-based strategy of search without any loss in the quality of plans or problem-solving coverage; and (3) For the strategy of model-based search, the “principle of locality” provides a productive basis for partitioning and organizing topological knowledge. We describe the design of a multistrategy navigation planner called Router that provides an experimental testbed for evaluation the three hypotheses. We also describe the embodiment of Router on a mobile robot called Stimpy for testing the first hypothesis. Experiments with Stimpy indicate that this hypothesis apparently is valid for hybrid robots in visually engineered spaces containing discrete pathways such as office buildings. In addition, two different kinds of simulation experiments with Router indicate that the second and the third hypotheses are only partially correct. Finally, we relate the evaluation methods and experimental designs with the research hypotheses.
Reasoning About Function in Reflective Systems

(Georgia Institute of Technology, 1993) Stroulia, Eleni ; Goel, Ashok K.

Functional models have been extensively investigated in the context of several problem-solving tasks such as device diagnosis and design. In this paper, we view problem solvers themselves as devices, use functional models to represent how they work, and subsequently employ these models for performance-driven reflective reasoning and learning. We represent the functioning of a problem solver as a structure-behavior-function model that specifies how the knowledge and reasoning of the problem solver results in the achievement of its goals. We view performance-driven learning as the task of redesigning the knowledge and reasoning of the problem solver. We use the structure-behavior-function model of the problem solver to monitor its reasoning, reflectively assign blame when it fails, and redesign its knowledge and reasoning. This paper describes an architecture for reflective model-based reasoning that is capable of a broad range of learning tasks. It also illustrates reflective model-based learning using examples from the Autognostic system, a reflective path planner.
An adaptive approach to qualitative modeling in design

(Georgia Institute of Technology, 1993) Goel, Ashok K.
Case-Based Decision Support: A Case Study in Architectural Design

(Georgia Institute of Technology, 1992) Zimring, Craig ; Pearce, Michael ; Goel, Ashok K. ; Kolodner, Janet L. ; Sentosa, Lucas Shindunata ; Billington, Richard