Georg Bartels

Constraint-based movement representation grounded in geometric features

Robots that are to master everyday manipulation tasks need both: The ability to reason about actions, objects and action effects, and the ability to perform sophisticated movement control. To bridge the gap between these two worlds, we consider the problem of connecting symbolic action representation with strategies from motion control engineering. We present a system using the task function approach [1] to define a common symbolic movement description language which defines motions as sets of symbolic constraints. We define these constraints using geometric features, like points, lines, and planes, grounding the description in the visual percepts of the robot. Additionally, we propose to assemble task functions by stacking 1-D feature functions, which leads to a modular movement specification. We evaluate and validate our approach on the task of flipping pancakes with a robot, showcasing the robustness and flexibility of the proposed movement representation.

Knowledge-based specification of robot motions

In many cases, the success of a manipulation action performed by a robot is determined by how it is executed and by how the robot moves during the action. Examples are tasks such as unscrewing a bolt, pouring liquids and flipping a pancake. This aspect is often abstracted away in AI planning and action languages that assume that an action is successful as long as all preconditions are fulfilled. In this paper we investigate how constraint-based motion representations used in robot control can be combined with a semantic knowledge base in order to let a robot reason about movements and to automatically generate executable motion descriptions that can be adapted to different robots, objects and tools.

Robotic agents capable of natural and safe physical interaction with human co-workers

Many future application scenarios of robotics envision robotic agents to be in close physical interaction with humans: On the factory floor, robotic agents shall support their human co-workers with the dull and health threatening parts of their jobs. In their homes, robotic agents shall enable people to stay independent, even if they have disabilities that require physical help in their daily life - a pressing need for our aging societies. A key requirement for such robotic agents is that they are safety-aware, that is, that they know when actions may hurt or threaten humans and actively refrain from performing them. Safe robot control systems are a current research focus in control theory. The control system designs, however, are a bit paranoid: programmers build “software fences” around people, effectively preventing physical interactions. To physically interact in a competent manner robotic agents have to reason about the task context, the human, and her intentions. In this paper, we propose to extend cognition-enabled robot control by introducing humans, physical interaction events, and safe movements as first class objects into the plan language. We show the power of the safety-aware control approach in a real-world scenario with a leading-edge autonomous manipulation platform. Finally, we share our experimental recordings through an online knowledge processing system, and invite the reader to explore the data with queries based on the concepts discussed in this paper.

Open robotics research using web-based knowledge services

In this paper we discuss how the combination of modern technologies in “big data” storage and management, knowledge representation and processing, cloud-based computation, and web technology can help the robotics community to establish and strengthen an open research discipline. We describe how we made the demonstrator of a EU project review openly available to the research community. Specifically, we recorded episodic memories with rich semantic annotations during a pizza preparation experiment in autonomous robot manipulation. Afterwards, we released them as an open knowledge base using the cloud- and web-based robot knowledge service OPENEASE. We discuss several ways on how this open data can be used to validate our experimental reports and to tackle novel challenging research problems.

Learning models for constraint-based motion parameterization from interactive physics-based simulation

For robotic agents to perform manipulation tasks in human environments at a human level or higher, they need to be able to relate the physical effects of their actions to how they are executing them; small variations in execution can have very different consequences. This paper proposes a framework for acquiring and applying action knowledge from naive user demonstrations in an interactive simulation environment under varying conditions. The framework combines a flexible constraint-based motion control approach with games-with-a-purpose-based learning using Random Forest Regression. The acquired action models are able to produce context-sensitive constraint-based motion descriptions to perform the learned action. A pouring experiment is conducted to test the feasibility of the suggested approach and shows the learned system can perform comparable to its human demonstrators.

Multidimensional Time-Series Shapelets Reliably Detect and Classify Contact Events in Force Measurements of Wiping Actions

The vision of service robots that autonomously manipulate objects as skillfully and flexibly as humans is still an open challenge. Findings from cognitive psychology suggest that the human brain structures manipulation actions along representations of contact events and their perceptually distinctive sensory signals. In this letter, we investigate how to reliably detect and classify contact events during robotic wiping actions. We present an algorithm that learns the distinct shapes of force measurements during contact events using multidimensional time-series shapelets. We evaluate our approach on a dataset consisting of 460 real-world robot wiping episodes that we collected using a table-mounted robot with a wrist-mounted force/torque sensor. Our approach shows good performance with tenfold cross validation yielding 97.5% precision and 99.3% recall, and can also be used for online contact event detection and classification.