Metrics for evaluating the quality of a human-robot interface are introduced. The autonomy of a robot is measured by its neglect time. The robot attention demand metric measures how much of the user’s attention is involved with instructing a robot. The free-time and fan-out metrics are two ways to measure this demand. Each of them leads to estimates of the interaction effort. Reducing interaction effort without diminishing task effectiveness is the goal of human-robot interaction design.

/pdf/metrics-for-evaluating-human-robot-interactions-6ics25y5tu.pdf

Metrics for Evaluating Human-Robot Interactions

This paper presents a summary of the space of commonly-used HCI prototyping methods (low-fidelity to high-fidelity) and asserts that with a better understanding of this space, HCI practitioners will be better equipped to direct scarce prototyping resources toward an effort likely to yield specific results It presents a set of five dimensions along which prototypes can be planned and characterized The paper then describes an analysis of this space performed by members of the NASA Ames Human-Computer Interaction Group when considering prototyping approaches for a new set of tools for Mars mission planning and scheduling tools A description is presented of a prototype that demonstrates design solutions that would have been particularly difficult to test given conventional low- or mid- fidelity prototyping methods The prototype created was "mixed-fidelity," that is, high-fidelity on some dimensions and low-fidelity on others The prototype is compared to a preexisting tool being redesigned and to a tool that has been developed using the prototype Experimental data are presented that show the prototype to be a good predictor of eventual user performance with the final application Given the relative cost of developing prototypes, it is critical to better characterize the space of fidelity in order to more precisely allocate design and development resources

Breaking the fidelity barrier: an examination of our current characterization of prototypes and an example of a mixed-fidelity success

While even STRIPS planners must search for plans of unbounded length, temporal planners must also cope with the fact that actions may start at any point in time. Most temporal planners cope with this challenge by restricting action start times to a small set of decision epochs, because this enables search to be carried out in state-space and leverages powerful state-based reachability heuristics, originally developed for classical planning. Indeed, decision-epoch planners won the International Planning Competition's Temporal Planning Track in 2002, 2004 and 2006.

However, decision-epoch planners have a largely unrecognized weakness: they are incomplete. In order to characterize the cause of incompleteness, we identify the notion of required concurrency, which separates expressive temporal action languages from simple ones. We show that decisionepoch planners are only complete for languages in the simpler class, and we prove that the simple class is 'equivalent' to STRIPS! Surprisingly, no problems with required concurrency have been included in the planning competitions. We conclude by designing a complete state-space temporal planning algorithm, which we hope will be able to achieve high performance by leveraging the heuristics that power decision epoch planners.

/pdf/when-is-temporal-planning-really-temporal-4mnuldxjas.pdf

When is temporal planning really temporal

In this paper we describe COLIN, a forward-chaining heuristic search planner, capable of reasoning with COntinuous LINear numeric change, in addition to the full temporal semantics of PDDL2.1. Through this work we make two advances to the state-of-the-art in terms of expressive reasoning capabilities of planners: the handling of continuous linear change, and the handling of duration-dependent effects in combination with duration inequalities, both of which require tightly coupled temporal and numeric reasoning during planning. COLIN combines FF-style forward chaining search, with the use of a Linear Program (LP) to check the consistency of the interacting temporal and numeric constraints at each state. The LP is used to compute bounds on the values of variables in each state, reducing the range of actions that need to be considered for application. In addition, we develop an extension of the Temporal Relaxed Planning Graph heuristic of CRIKEY3, to support reasoning directly with continuous change. We extend the range of task variables considered to be suitable candidates for specifying the gradient of the continuous numeric change effected by an action. Finally, we explore the potential for employing mixed integer programming as a tool for optimising the timestamps of the actions in the plan, once a solution has been found. To support this, we further contribute a selection of extended benchmark domains that include continuous numeric effects. We present results for COLIN that demonstrate its scalability on a range of benchmarks, and compare to existing state-of-the-art planners.

/pdf/colin-planning-with-continuous-linear-numeric-change-3vap4xcmnn.pdf

COLIN: planning with continuous linear numeric change

When AI systems interact with humans in the loop, they are often called on to provide explanations for their plans and behavior. Past work on plan explanations primarily involved the AI system explaining the correctness of its plan and the rationale for its decision in terms of its own model. Such soliloquy is wholly inadequate in most realistic scenarios where the humans have domain and task models that differ significantly from that used by the AI system. We posit that the explanations are best studied in light of these differing models. In particular, we show how explanation can be seen as a "model reconciliation problem" (MRP), where the AI system in effect suggests changes to the human's model, so as to make its plan be optimal with respect to that changed human model. We will study the properties of such explanations, present algorithms for automatically computing them, and evaluate the performance of the algorithms.

Plan Explanations as Model Reconciliation: Moving Beyond Explanation as Soliloquy

The Mars Exploration Rover mission is one of NASA's most ambitious science missions to date. Launched in the summer of 2003, each rover carries instruments for conducting remote and in site observations to elucidate the planet's past climate, water activity, and habitability. Science is MER's primary driver, so making best use of the scientific instruments, within the available resources, is a crucial aspect of the mission. To address this criticality, the MER project team selected MAPGEN (mixed initiative activity plan generator) as an activity-planning tool. MAPGEN combines two existing systems, each with a strong heritage: the APGEN activity-planning tool from the Jet Propulsion Laboratory and the Europa planning and scheduling system from NASA Ames Research Center. We discuss the issues arising from combining these tools in this mission's context. MAPGEN is the first AI-based system to control a space platform on another planet's surface.

https://ti.arc.nasa.gov/m/pub-archive/542h/0542%20(Ai-Chang)%20v3.pdf

MAPGEN: mixed-initiative planning and scheduling for the Mars Exploration Rover mission

The Mars Exploration Rovers Mars '03 mission is one of NASA's most ambitious science missions to date. The rovers will be launched in the summer of 2003 with each rover carrying instruments to conduct remote and in-situ observation to elucidate the planet's past climate, water activity, and habitability. Science is the primary driver of MER and, as a consequence, making best use of the scientific instruments, within the available resources, is a crucial aspect of the mission. To address this critically, the MER project has selected MAPGEN (Mixed-Initiative Activity Plan GENerator) as an activity planning tool. MAPGEN combines two exiting systems, each with a strong heritage: APGEN the Activity Planning tool from the Jet Propulsion Laboratory and the Europs Planning/Scheduling system from NASA Ames Research Center. This paper discusses the issues arising from combining these tools in the context of this mission.

/pdf/mapgen-mixed-initiative-planning-and-scheduling-for-the-mars-1xyj2krahs.pdf

MAPGEN : mixed initiative planning and scheduling for the Mars '03 MER mission

Flight controllers manage the orientation and modes of eight large solar arrays that power the International Space Station (ISS). The task requires generating plans that balance complex constraints and preferences. These considerations include context-dependent constraints on viable solar array configurations, temporal limits on transitions between configurations, and preferences on which considerations have priority. The Solar Array Constraint Engine (SACE) treats this operations planning problem as a sequence of tractable constrained optimization problems. SACE uses constraint management and automated planning capabilities to reason about the constraints, to find optimal array configurations subject to these constraints and solution preferences, and to automatically generate solar array operations plans. SACE further provides flight controllers with real-time situational awareness and what-if analysis capabilities. SACE is built on the Extensible Universal Remote Operations Planning Architecture (EUROPA) model-based planning system. EUROPA facilitated SACE development by providing model-based planning, built-in constraint reasoning capability, and extensibility. This article formulates the planning problem, explains how EUROPA solves the problem, and provides performance statistics from several planning scenarios. SACE reduces a highly manual process that takes weeks to an automated process that takes tens of minutes.

Mitchell Ai-Chang

Papers

MAPGEN: mixed-initiative planning and scheduling for the Mars Exploration Rover mission

MAPGEN : mixed initiative planning and scheduling for the Mars '03 MER mission

Planning solar array operations on the international space station