Refine
H-BRS Bibliography
- yes (87) (remove)
Departments, institutes and facilities
- Fachbereich Informatik (38)
- Fachbereich Ingenieurwissenschaften und Kommunikation (21)
- Institut für Technik, Ressourcenschonung und Energieeffizienz (TREE) (17)
- Fachbereich Wirtschaftswissenschaften (15)
- Institut für Verbraucherinformatik (IVI) (14)
- Institut für Cyber Security & Privacy (ICSP) (13)
- Internationales Zentrum für Nachhaltige Entwicklung (IZNE) (8)
- Fachbereich Sozialpolitik und Soziale Sicherung (5)
- Institute of Visual Computing (IVC) (4)
- Zentrum für Innovation und Entwicklung in der Lehre (ZIEL) (3)
Document Type
- Conference Object (87) (remove)
Year of publication
- 2021 (87) (remove)
Keywords
- Augmented Reality (3)
- Big Data Analysis (2)
- Cognitive robot control (2)
- Explainable robotics (2)
- Learning from experience (2)
- Usable Privacy (2)
- Usable Security (2)
- AES (1)
- AR design (1)
- AR development (1)
Robot Action Diagnosis and Experience Correction by Falsifying Parameterised Execution Models
(2021)
When faced with an execution failure, an intelligent robot should be able to identify the likely reasons for the failure and adapt its execution policy accordingly. This paper addresses the question of how to utilise knowledge about the execution process, expressed in terms of learned constraints, in order to direct the diagnosis and experience acquisition process. In particular, we present two methods for creating a synergy between failure diagnosis and execution model learning. We first propose a method for diagnosing execution failures of parameterised action execution models, which searches for action parameters that violate a learned precondition model. We then develop a strategy that uses the results of the diagnosis process for generating synthetic data that are more likely to lead to successful execution, thereby increasing the set of available experiences to learn from. The diagnosis and experience correction methods are evaluated for the problem of handle grasping, such that we experimentally demonstrate the effectiveness of the diagnosis algorithm and show that corrected failed experiences can contribute towards improving the execution success of a robot.
When an autonomous robot learns how to execute actions, it is of interest to know if and when the execution policy can be generalised to variations of the learning scenarios. This can inform the robot about the necessity of additional learning, as using incomplete or unsuitable policies can lead to execution failures. Generalisation is particularly relevant when a robot has to deal with a large variety of objects and in different contexts. In this paper, we propose and analyse a strategy for generalising parameterised execution models of manipulation actions over different objects based on an object ontology. In particular, a robot transfers a known execution model to objects of related classes according to the ontology, but only if there is no other evidence that the model may be unsuitable. This allows using ontological knowledge as prior information that is then refined by the robot’s own experiences. We verify our algorithm for two actions – grasping and stowing everyday objects – such that we show that the robot can deduce cases in which an existing policy can generalise to other objects and when additional execution knowledge has to be acquired.
Target meaning representations for semantic parsing tasks are often based on programming or query languages, such as SQL, and can be formalized by a context-free grammar. Assuming a priori knowledge of the target domain, such grammars can be exploited to enforce syntactical constraints when predicting logical forms. To that end, we assess how syntactical parsers can be integrated into modern encoder-decoder frameworks. Specifically, we implement an attentional SEQ2SEQ model that uses an LR parser to maintain syntactically valid sequences throughout the decoding procedure. Compared to other approaches to grammar-guided decoding that modify the underlying neural network architecture or attempt to derive full parse trees, our approach is conceptually simpler, adds less computational overhead during inference and integrates seamlessly with current SEQ2SEQ frameworks. We present preliminary evaluation results against a recurrent SEQ2SEQ baseline on GEOQUERY and ATIS and demonstrate improved performance while enforcing grammatical constraints.
Property-Based Testing in Simulation for Verifying Robot Action Execution in Tabletop Manipulation
(2021)
An important prerequisite for the reliability and robustness of a service robot is ensuring the robot’s correct behavior when it performs various tasks of interest. Extensive testing is one established approach for ensuring behavioural correctness; this becomes even more important with the integration of learning-based methods into robot software architectures, as there are often no theoretical guarantees about the performance of such methods in varying scenarios. In this paper, we aim towards evaluating the correctness of robot behaviors in tabletop manipulation through automatic generation of simulated test scenarios in which a robot assesses its performance using property-based testing. In particular, key properties of interest for various robot actions are encoded in an action ontology and are then verified and validated within a simulated environment. We evaluate our framework with a Toyota Human Support Robot (HSR) which is tested in a Gazebo simulation. We show that our framework can correctly and consistently identify various failed actions in a variety of randomised tabletop manipulation scenarios, in addition to providing deeper insights into the type and location of failures for each designed property.
Execution monitoring is essential for robots to detect and respond to failures. Since it is impossible to enumerate all failures for a given task, we learn from successful executions of the task to detect visual anomalies during runtime. Our method learns to predict the motions that occur during the nominal execution of a task, including camera and robot body motion. A probabilistic U-Net architecture is used to learn to predict optical flow, and the robot's kinematics and 3D model are used to model camera and body motion. The errors between the observed and predicted motion are used to calculate an anomaly score. We evaluate our method on a dataset of a robot placing a book on a shelf, which includes anomalies such as falling books, camera occlusions, and robot disturbances. We find that modeling camera and body motion, in addition to the learning-based optical flow prediction, results in an improvement of the area under the receiver operating characteristic curve from 0.752 to 0.804, and the area under the precision-recall curve from 0.467 to 0.549.
Over the last decades, different kinds of design guides have been created to maintain consistency and usability in interactive system development. However, in the case of spatial applications, practitioners from research and industry either have difficulty finding them or perceive such guides as lacking relevance, practicability, and applicability. This paper presents the current state of scientific research and industry practice by investigating currently used design recommendations for mixed reality (MR) system development. We analyzed and compared 875 design recommendations for MR applications elicited from 89 scientific papers and documentation from six industry practitioners in a literature review. In doing so, we identified differences regarding four key topics: Focus on unique MR design challenges, abstraction regarding devices and ecosystems, level of detail and abstraction of content, and covered topics. Based on that,we contribute to the MR design research by providing three factors for perceived irrelevance and six main implications for design recommendations that are applicable in scientific and industry practice.
Augmented/Virtual Reality (AR/VR) is still a fragmented space to design for due to the rapidly evolving hardware, the interdisciplinarity of teams, and a lack of standards and best practices. We interviewed 26 professional AR/VR designers and developers to shed light on their tasks, approaches, tools, and challenges. Based on their work and the artifacts they generated, we found that AR/VR application creators fulfill four roles: concept developers, interaction designers, content authors, and technical developers. One person often incorporates multiple roles and faces a variety of challenges during the design process from the initial contextual analysis to the deployment. From analysis of their tool sets, methods, and artifacts, we describe critical key challenges. Finally, we discuss the importance of prototyping for the communication in AR/VR development teams and highlight design implications for future tools to create a more usable AR/VR tool chain.
New cars are increasingly "connected" by default. Since not having a car is not an option for many people, understanding the privacy implications of driving connected cars and using their data-based services is an even more pressing issue than for expendable consumer products. While risk-based approaches to privacy are well established in law, they have only begun to gain traction in HCI. These approaches are understood not only to increase acceptance but also to help consumers make choices that meet their needs. To the best of our knowledge, perceived risks in the context of connected cars have not been studied before. To address this gap, our study reports on the analysis of a survey with 18 open-ended questions distributed to 1,000 households in a medium-sized German city. Our findings provide qualitative insights into existing attitudes and use cases of connected car features and, most importantly, a list of perceived risks themselves. Taking the perspective of consumers, we argue that these can help inform consumers about data use in connected cars in a user-friendly way. Finally, we show how these risks fit into and extend existing risk taxonomies from other contexts with a stronger social perspective on risks of data use.
Since stationary self-checkout is widely introduced and well understood, previous research barely examined newer generations of smartphone-based Scan&Go. Especially from a design perspective, we know little about the factors contributing to the adoption of Scan&Go solutions and how design enables consumers to take full advantage of this development rather than being burdened with using complex and unenjoyable systems. To understand the influencing factors and the design from a consumer perspective, we conducted a mixed-methods study where we triangulated data of an online survey with 103 participants and a qualitative study with 20 participants. Based on the results, our study presents a refined and nuanced understanding of technology as well as infrastructure-related factors that influence adoption. Moreover, we present several implications for designing and implementing of Scan&Go in retail environments.