those of the author and should not be interpreted as representing the official policies, either expressed or implied, of any sponsoring institution, the U.S. government or any other entity. Keywords: robotics The problem of learning a policy, a task representation mapping from world states to actions, lies at the heart of many robotic applications. One approach to acquiring a task policy is learning from demonstration, an interactive technique in which a robot learns a policy based on example state to action mappings provided by a human teacher. This thesis introduces Confidence-Based Autonomy, a mixed-initiative single robot demonstration learning algorithm that enables the robot and teacher to jointly control the learning process and selection of demonstration training data. The robot to identifies the need for and requests demonstrations for specific parts of the state space based on confidence thresholds characterizing the uncertainty of the learned policy. The robot’s demonstration requests are complemented by the teacher’s ability to provide supplementary corrective demonstrations in error cases. An additional algorithmic component enables choices between multiple equally

Existing approaches to multirobot coordination separate scheduling and task allocation, but finding the optimal schedule with joint tasks and spatial constraints requires robots to simultaneously solve the scheduling, task allocation, and path planning problems. We present a formal description of the multirobot joint task allocation problem with heterogeneous capabilities and spatial constraints and an instantiation of the problem for the search and rescue domain. We introduce a novel declarative framework for modeling the problem as a mixed integer linear programming (MILP) problem and present a centralized anytime algorithm with error bounds. We demonstrate that our algorithm can outperform standard MILP solving techniques, greedy heuristics, and a market based approach which separates scheduling and task allocation.

comparative study of task assignment and path

Sub area code

The Multirobot Task Allocation (MRTA) paradigm is widely used in multirobot cooperation schemes, e.g. for observation, surveillance and tracking missions. Market-based approaches have yielded effective distributed solutions for such missions, showing the ability to manage heterogeneous, dynamic and robust robot teams. Two major challenges remain however poorly tackled: the management of inter-robot and inter-task communication constraints, and the use of a rich task formalism to model complex missions. This paper presents our investigations to treat these two aspects. The inter-robot and inter-task communication constraints are explicitly handled in the task allocation process, through simple geometric models and thanks to temporal scheduling skills. Rich tasks are allocated using a treebased task formalism that allows to treat complex missions with task ordering. Current work has shown it to be able to handle more complex tasks and to give better solution than MRTA systems working on simple task structures. In our work we will try to go further in this investigation.

Abstract — Humanoid robots working alongside humans in everyday environments is a long standing goal of the robotics community. To achieve this goal, methods for developing new robot behaviors that are intuitive and accessible to non-programmers are required. In this paper, we present a demonstration-based method for teaching distributed autonomous robots to coordinate their actions and perform collaborative multi-robot tasks. Within the presented framework, each robot learns an individual policy from teacher demonstrations using a confidence-based algorithm. Based on this learning approach, we contribute three techniques for teaching multi-robot coordination using different information sharing strategies. We evaluate and compare these approaches by teaching two Sony QRIO humanoid robots to perform three collaborative ball sorting tasks. I.