Hierarchy in Flux: Interfacing Robots
Download as PPTX, PDF1 like819 views
The document discusses the use of tele-operated robots for retrieving airplane black boxes, focusing on a sociotechnical system with a well-defined hierarchical structure that minimizes interaction among specialized roles. It emphasizes the importance of human factors, cognitive contexts, and emergent uses of technology in designing efficient interfaces and systems. Additionally, it explores the challenge of transforming rigid systems into adaptable ones while considering the dynamics of interface design and user engagement.
Hierarchy in Flux: Interfacing Robots
- 2. Scenario: Retrieve a airplane Blackbox Tele-operated robot with toolkit (controlled by umbilicus 6sec delay) Robot operators in control room (video, telemetry, sensor arrays) “Dry” testing environment
- 3. Sociotechnical System Tele-operated robot Robot operators Interface Dry-testing Leaders
- 4. • Weak Emergence “predictable collective action” • Strong Emergence “effects you could not anticipate or deduce” Emergence: higher scale effects
- 5. Maximize Weak Emergence Minimize (eliminate) strong emergence Limit interaction between parts of the system Engineering Emergence
- 8. Every role is specialized Every specialist is focused on one small set of tasks Every task is clearly defined Inputs and outputs only go up or down one level It’s a rigid and well-defined hierarchy that minimizes interaction and organizes the flow of communication and control It is engineered not self-organizing Why does this work?
- 9. 1 – Human Factors Emphasizes human-machine coupling Treats user as blackbox with inputs and outputs 2 – Cognitivist Emphasizes the work/task context Supports the user as an intentional agent 3 – Phenomenological Emphasizes emergent uses of technology Understands the user as a source of meanings Wave Theory of HCI
- 10. Decomposes activity into “Activity, Action, Operation” hierarchy. These closely map to Knowledge, Rules, and Skills, respectively. Activity Theory
- 11. Reconfiguring the Social Hierarchy How do we turn the rigid engineered system into a lightweight adaptable one? Parsimony with variety Co-locate personnel (from iceberg to ice cube) Redundancy and variability of roles (flexibility of interface) Automate skills (build them into the robot) Dry-testing and modeling
- 12. What does this have to do with interface design? Understanding context is important, but there is a problem with the unit of analysis (level of description) We’ve designed a context but not an interface More like a waterfall than co-evolution When we begin to look at the design of the interface itself a new set of dynamics begin to dominate Perception, reasoning, situated-ness, communication
- 13. How do we bridge the gap? ORGANIZATION Sociotechnical System Technologies and UI Sociotechnical System INTERFACE (UI) DATA and USERS Theories & Methods? Representation & Feedback? Environment System Components Environment System Components
Editor's Notes
- #2: Can’t show you the interfaces, can’t show you the data, Can discuss the qualitative hypothesis-forming work. So I will end with a question that we’ve uncovered that I hope we can discuss a little bit.
- #4: Various configurations of people and technology all organized to achieve a goal The purpose of this system is to create an emergent result, to retrieve the blackbox
- #5: The STS relies on emergence to meet its goal, that is the different parts of the system all interact over time to produce the emergent effect of higher scale action. Weak emergence is good because its people working together to achieve higher aims that makes executing a mission possible Strong emergence will ruin your day.
- #7: Every numbered position is a “discipline” responsible for some aspect of the mission, monitoring electrical systems, etc…
- #10: Important to recognize that this is a really accurate and totally inaccurate picture of HCI as a field