Jason Robert Chen

Programing by Demonstration: Coping with Suboptimal Teaching Actions

The difficulty associated with programing existing robots is one of the main impediments to them finding application in domestic environments such as the home. A promising method for simplifying robot programing is Programing by Demonstration (PbD). Here, an end user can provide a demonstration of the task to be programed, with a PbD “interface” interpreting the demonstration in order to determine low-level control details for the robot. A key aspect of the interpretation process is to make it robust to the noise typically included in a demonstration by the human. In this paper we present a method to help identify and eliminate any noise present in the demonstration. Our method involves two steps. The first step uses the demonstration to build up a partial knowledge of the geometry present in the task. Statistical regression analysis is used on demonstrated trajectories to determine equations describing curved surfaces in configuration space. The second step in our method uses the geometric information obtained in the first step to determine if there are more optimal paths than those demonstrated for completing the task. If there are, our method proposes these as the appropriate control commands for the robot. We show the validity of our approach by presenting successful experiments on a realistic household-type task—changing rolls on a paper roll holder.

Constructing Task-Level Assembly Strategies in Robot Programming by Demonstration

Programming by demonstration (PbD) is a technique for programming robots that holds much promise in making robots more accessible to ordinary, non-technical users. However, a well-known difficulty with the method is that a human will often demonstrate the task to be programmed inconsistently or even erroneously, leading to the inclusion of what is essentially noise in the demonstration. A number of techniques exist in the literature for filtering out this type of noise; however, most focus on very low level control command details. In this paper, we propose a new, complementary direction of research. We take a “task-level” view of the demonstration, and note that noise can exist at this level also. We propose a framework, based on a hybrid dynamic system modeling approach, to select the most optimal, task-level execution strategies that were demonstrated. We apply our framework to a real household task of inserting the compressible spindle of a paper towel holder into its supports. We conduct experiments to show that significant improvements in robot performance of the task can be achieved by a PbD regime that includes our method.

Making clustering in delay-vector space meaningful

Sequential time series clustering is a technique used to extract important features from time series data. The method can be shown to be the process of clustering in the delay-vector space formalism used in the Dynamical Systems literature. Recently, the startling claim was made that sequential time series clustering is meaningless. This has important consequences for a significant amount of work in the literature, since such a claim invalidates these work’s contribution. In this paper, we show that sequential time series clustering is not meaningless, and that the problem highlighted in these works stem from their use of the Euclidean distance metric as the distance measure in the delay-vector space. As a solution, we consider quite a general class of time series, and propose a regime based on two types of similarity that can exist between delay vectors, giving rise naturally to an alternative distance measure to Euclidean distance in the delay-vector space. We show that, using this alternative distance measure, sequential time series clustering can indeed be meaningful. We repeat a key experiment in the work on which the “meaningless” claim was based, and show that our method leads to a successful clustering outcome.

Programming by demonstration - constructing task level plans in hybrid dynamic framework

This paper presents a novel approach to constructing a task level plan for an assembly task from demonstration. A hybrid dynamic system is chosen as an attractive way to model assembly tasks. We propose a framework where the mappings required within the hybrid dynamic system model are extracted from a demonstration. We extract the event path planner mapping, which determines a task level plan. Demonstrated paths are broken down into their base elements, called transitions. A path plan is then constructed from transitions that were well demonstrated. We test the approach by having a robot execute an assembly task using constructed task plans. The framework produces excellent results because: 1) the robot can perform selected paths better than when it simply copies the demonstrator, and 2) it allows flexible path selection so the robot can be given a disposition in how it executes the task.

Programming by demonstration: removing sub-optimal actions in a partially known configuration space

Programming by demonstration is a promising approach to automatic robot programming, however, methods are required to remove suboptimal actions that can be demonstrated by end users. In this paper we use the partial knowledge of configuration space (C-space) derived in the previous work by Chen et al. (2000) to remove suboptimal actions from a demonstration. Our idea is to use demonstrated paths to predict what regions in C-space are obstacle free. Suboptimal actions in a demonstration are then avoided by planning alternative actions that pass through the obstacle free regions. Experimental results show the validity of the approach. A demonstrated path containing significant sub-optimality was converted by the approach into a short, efficient path suitable for execution by the robot.

Robot programming by demonstration-selecting optimal event paths

Presents a framework for robot programming by human demonstration. The framework builds a high level robot controller using information extracted from the demonstration. The high level robot controller is broken down into three component parts, each fulfilling a different function during execution. The paper focuses on the construction of an event path planner, which determines the optimal event path of a task. The approach varies the optimal path according to what characteristics of the demonstrations are stressed, thus giving the robot a selected disposition. The approach was implemented on a simple navigational task. The event path planner selected appropriate paths and could change its selection according to what characteristics were desired in the selected path.

Generating a configuration space representation for assembly tasks from demonstration

Removing suboptimal actions that can exist in a demonstration is a key problem to be solved in robot programming by demonstration. In this paper we present the first step of an approach for solving this problem. We present how the configuration space (C-space) of a task can be derived from demonstration. A demonstration traces out paths on a number of C-surfaces in C-space. The idea is to use statistical regression analysis on data from these paths to determine the unknown equation parameters of a C-surface. Experimental results show the validity of the approach. Accurate parameter estimates were obtained so long as a sufficiently rich set of demonstrated paths existed on the C-surface. The approach has the advantage that it tends to provide accurate parameter estimates for C-surfaces where they were most needed; that is, for C-surfaces (i) critical to task completion, and (ii) whose paths contained suboptimal actions.

Symbol Statistics for Concept Formation in AI Agents

High level conceptual thought seems to be at the basis of the impressive human cognitive ability. Classical top-down (Logic based) and bottom-up (Connectionist) approaches to the problem have had limited success to date. We identify a small body of work that represents a different approach to AI. We call this work the Bottom Up Symbolic (BUS) approach and present a new BUS method to concept construction. The main novelty of our work is that we apply statistical methods in the concept construction process. Our findings here suggest that such methods are necessary since a symbolic description of the true agent-environment interaction dynamics is often hidden among a background of non-representative descriptions, especially if data from unconstrained real-world experiments is considered. We consider such data (from a mobile robot randomly roaming an office environment) and show how our method can correctly grow a set of true concepts from the data.

Building concepts for AI agents using information theoretic Co-clustering

High level conceptual thought seems to be at the basis of the impressive human cognitive ability, and AI researchers aim to replicate this ability in artificial agents. Classical top-down (Logic based) and bottom-up (Connectionist) approaches to the problem have had limited success to date. We review a small body of work that represents a different approach to AI. We call this work the Bottom Up Symbolic (BUS) approach and present a new BUS method to concept construction. While valid concepts have been constructed using previous methods under this approach, we show in this paper that the one-sided clustering methods generally used there may fail to uncover valid concepts even when they clearly exist. We show that by using a Co-clustering algorithm that searches for an optimal partitioning based on the Mutual Information between the category and consequent components of a concept, the concept formation outcome is improved. We test our approach on data from experiments using a real mobile robot operating in the real world, and show that our Co-clustering based approach leads to significant performance improvement compared to previous approaches.