Model-Lite Planning: Diverse Multi-Option Plans & Dynamic Objective Functions

Knowledge acquisition is one major bottle-neck in using planning systems. Model-lite planning reduces this burden by placing responsibility on the planning system to cope with partially specified models. For example, eliciting the planning objective can be difficult in applications where it is necessary to reason about multiple plan metrics, such as cost, time, risk, human life, etc. Traditional approaches, often require a (sometimes subjective) combination of these objectives into a single optimization metric. For example, decision theoretic planners combine plan cost and probability of goal satisfaction into a single reward metric. However, users may not know how to combine their metrics into a single objective without first exploring several diverse plan options. To avoid premature objective function commitments at plan synthesis time (and even plan execution time), we develop the notion of multi-option plans. Much like conditional plans that branch to deal with execution-time observations, multi-option plans branch to deal with execution-time assessments of plan objectives. That is, a multi-option plan is a compact representation of the diverse Pareto set of plans, where at each step the user can execute one of several non-dominated options. We formulate multi-option planning within the context of conditional probabilistic planning, where plans satisfy the goal with different probabilities and costs. Our approach is based on multi-objective dynamic programming in state space, where each plan node maintains a set of nondominated sub-plan options, that are each a conditional plan.