Hybrid Conditional Planning for Service Robotics

Ahmed Nouman
CogRobo Lab., Mechantronics Engineering, PhD Dissertation, 2018

Thesis Jury

Assoc. Prof Dr. Volkan Patoğlu (Thesis Advisor), Assoc. Prof Dr. Esra Erdem (Co-Advisor), Assoc. Prof. Dr. Güllü Kızıltaş Şendur, Assist. Prof. Dr. Kamer Kaya, Assist. Prof. Dr. Orkunt Sabuncu, Assist. Prof. Dr. Reyyan Yeniterzi

Date & Time:  7th, December 2018 –  12:40 PM

Place:  G029

Keywords :  Conditional Planning, Task Planning, Motion Planning, Hybrid Planning, Plan Execution Monitoring

Abstract

Conditional planning enables planning for the sensing actions and their possible outcomes in addition to actuation actions, and allows for addressing uncertainties due to partial observability at the time of offline planning. Therefore, the plans (called conditional plans) computed by conditional planners can be viewed as trees of (deterministic) actuation actions and (nondeterministic) sensing actions. Hybrid conditional planning extends conditional planning further by integrating low-level feasibility checks into executability conditions of actuation actions in conditional plans. We introduce a novel hybrid conditional planning method, which extends hybrid sequential planning with nondeterministic sensing actions and utilizes this extension to compute branches of a conditional plan in parallel. We evaluate this method in a service robotics domain, by means of a set of experiments over dynamic simulations, from the perspectives of computational efficiency and plan quality.

We further present two extensions to hybrid conditional planning: 1) Hybrid conditional partial planning that integrates probablistic framework for outcomes of sensing actions to compute efficient plans 2) Hybrid conditional reactive planning that alternates between a reaction hybrid conditional planning and execution cycle to reduce offline planning time and generate feasible plans.