Synthesizing manipulation sequences for under-specified tasks using unrolled Markov Random Fields

Many tasks in human environments require performing a sequence of navigation and manipulation steps involving objects. In unstructured human environments, the location and configuration of the objects involved often change in unpredictable ways. This requires a high-level planning strategy that is robust and flexible in an uncertain environment. We propose a novel dynamic planning strategy, which can be trained from a set of example sequences. High level tasks are expressed as a sequence of primitive actions or controllers (with appropriate parameters). Our score function, based on Markov Random Field (MRF), captures the relations between environment, controllers, and their arguments. By expressing the environment using sets of attributes, the approach generalizes well to unseen scenarios. We train the parameters of our MRF using a maximum margin learning method. We provide a detailed empirical validation of our overall framework demonstrating successful plan strategies for a variety of tasks. 1

Authors

• Jaeyong Sung
• Bart Selman
• Ashutosh Saxena

Keywords

• Robots
• Planning
• Vectors
• Liquids
• Markov random fields
• Sequential analysis
• Navigation
• Markov Random Field
• Scoring Function
• Sequential Steps
• Control Sequence
• Human Environment
• Task Environment
• Maximum Margin
• Unstructured Environments
• High-level Tasks
• Dynamic Planning
• Time Step
• Object Detection
• Weight Vector
• Multiple Objects
• Object Of Interest
• Path Planning
• Error Detection
• Manipulation Tasks
• Goal State
• Inverse Reinforcement Learning
• Explicit Labels
• Object C
• Object Labels
• Baseline Algorithms
• Reward Function
• Discrete Time Steps
• Correct Sequence
• Task Planning
• Navigation Control