Affordances - i.e. possibilities for action that an environment or objects in it provide - are important for robots operating in human environments to perceive. Existing approaches train such capabilities on annotated static images or shapes. This work presents a novel dataset for affordance learning of common household tasks. Unlike previous approaches, our dataset consists of video sequences demonstrating the tasks from first- and third-person perspectives, along with metadata about the affordances that are manifested in the task, and is aimed towards training perception systems to recognize affordance manifestations. The demonstrations were collected from several participants and in total record about seven hours of human activity. The variety of task performances also allows studying preparatory maneuvers that people may perform for a task, such as how they arrange their task space, which is also relevant for collaborative service robots.

In robotics, the effective integration of environ-mental data into actionable knowledge remains a significant challenge due to the variety and incompatibility of data formats commonly used in scene descriptions, such as MJCF, URDF, and SDF. This paper presents a novel approach that addresses these challenges by developing a unified scene graph model that standardizes these varied formats into the Universal Scene Description (USD) format. This standardization facilitates the integration of these scene graphs with robot ontologies through semantic reporting, enabling the translation of complex environmental data into actionable knowledge essential for cognitive robotic control. We evaluated our approach by converting procedural 3D environments into USD format, which is then annotated semantically and translated into a knowledge graph to effectively answer competency questions, demonstrating its utility for real-time robotic decision-making. Additionally, we developed a web-based visualization tool to support the semantic mapping process, providing users with an intuitive interface to manage the 3D environment.

This study explores which factors of the visual design of a robot may influence how humans would place it in a collaborative cooking scenario and how these features may influence task delegation. Human participants were placed in a Virtual Reality (VR) environment and asked to set up a kitchen for cooking alongside a robot companion while considering the robot's morphology. We collected multimodal data for the arrangements created by the participants, transcripts of their think-aloud as they were performing the task, and transcripts of their answers to structured post-task questionnaires. Based on analyzing this data, we formulate several hypotheses: humans prefer to collaborate with biomorphic robots; human beliefs about the sensory capabilities of robots are less influenced by the morphology of the robot than beliefs about action capabilities; and humans will implement fewer avoidance strategies when sharing space with gracile robots. We intend to verify these hypotheses in follow-up studies.

This paper introduces a novel benchmark that has been designed as a test bed for evaluating whether artificial agents are able to understand how to perform everyday activities, with a focus on the cooking domain. Understanding how to cook recipes is a highly challenging endeavour due to the underspecified and grounded nature of recipe texts, combined with the fact that recipe execution is a knowledge-intensive and precise activity. The benchmark comprises a corpus of recipes, a procedural semantic representation language of cooking actions, qualitative and quantitative kitchen simulators, and a standardised evaluation procedure. Concretely, the benchmark task consists in mapping a recipe formulated in natural language to a set of cooking actions that is precise enough to be executed in the simulated kitchen and yields the desired dish. To overcome the challenges inherent to recipe execution, this mapping process needs to incorporate reasoning over the recipe text, the state of the simulated kitchen environment, common-sense knowledge, knowledge of the cooking domain, and the action space of a virtual or robotic chef. This benchmark thereby addresses the growing interest in human-centric systems that combine natural language processing and situated reasoning to perform everyday activities.

Situationally-aware artificial agents operating with competence in natural environments face several challenges: spatial awareness, object affordance detection, dynamic changes and unpredictability. A critical challenge is the agent’s ability to identify and monitor environmental elements pertinent to its objectives. Our research introduces a neurosymbolic modular architecture for reactive robotics. Our system combines a neural component performing object recognition over the environment and image processing techniques such as optical flow, with symbolic representation and reasoning. The reasoning system is grounded in the embodied cognition paradigm, via integrating image schematic knowledge in an ontological structure. The ontology is operatively used to create queries for the perception system, decide on actions, and infer entities’ capabilities derived from perceptual data. The combination of reasoning and image processing allows the agent to focus its perception for normal operation as well as discover new concepts for parts of objects involved in particular interactions. The discovered concepts allow the robot to autonomously acquire training data and adjust its subsymbolic perception to recognize the parts, as well as making planning for more complex tasks feasible by focusing search on those relevant object parts. We demonstrate our approach in a simulated world, in which an agent learns to recognize parts of objects involved in support relations. While the agent has no concept of handle initially, by observing examples of supported objects hanging from a hook it learns to recognize the parts involved in establishing support and becomes able to plan the establishment/destruction of the support relation. This underscores the agent’s capability to expand its knowledge through observation in a systematic way, and illustrates the potential of combining deep reasoning with reactive robotics in dynamic settings.

Robots performing human-scale manipulation tasks require an extensive amount of knowledge about their surroundings in order to perform their actions competently and human-like. In this work, we investigate the use of virtual reality technology as an implementation for robot environment modeling, and present a technique for translating scene graphs into knowledge bases. To this end, we take advantage of the Universal Scene Description (USD) format which is an emerging standard for the authoring, visualization and simulation of complex environments. We investigate the conversion of USD-based environment models into Knowledge Graph (KG) representations that facilitate semantic querying and integration with additional knowledge sources. The contributions of the paper are validated through an application scenario in the service robotics domain.

The Collaborative Research Center for Everyday Activity Science & Engineering (CRC EASE) aims to enable robots to perform environmental interaction tasks with close to human capacity. It therefore employs a shared ontology to model the activity of both kinds of agents, empowering robots to learn from human experiences. To properly describe these human experiences, the ontology will strongly benefit from incorporating characteristics of neuronal information processing which are not accessible from a behavioral perspective alone. We, therefore, propose the analysis of human neuroimaging data for evaluation and validation of concepts and events defined in the ontology model underlying most of the CRC projects. In an exploratory analysis, we employed an Independent Component Analysis (ICA) on functional Magnetic Resonance Imaging (fMRI) data from participants who were presented with the same complex video stimuli of activities as robotic and human agents in different environments and contexts. We then correlated the activity patterns of brain networks represented by derived components with timings of annotated event categories as defined by the ontology model. The present results demonstrate a subset of common networks with stable correlations and specificity towards particular event classes and groups, associated with environmental and contextual factors. These neuronal characteristics will open up avenues for adapting the ontology model to be more consistent with human information processing.

We present the Meta-Ontology for Introspection (MOI): Inspired by fundamental processes of the human mind, cognitive architectures (CAs) explore ever more methods to leverage metacognition. Still, an ontological model to trace metacognitive experiences for learning or as input for metacognitive control routines has yet to be developed. Based on a review of existing standards, we formally identify the relevant scope in the form of Competency Questions (CQs) and extend SOMA, a well-established formal ontology initially designed to interpret episodic memories of a robotic CA. The resulting MOI can model a CA’s software and capabilities of single components, trace information processing and inter-component communication, label self-lived mental events, and capture causal relationships. We evaluate MOI via the CQs and exemplarily demonstrate its reasoning capabilities.

Abstract Human beings and other biological agents appear driven by curiosity to explore the affordances of their environments. Such exploration is its own reward – children have fun when playing – but it probably also serves the practical purpose of learning theories with which to predict outcomes of actions. Cognitive robots however have yet to match the performance of human beings at learning and reusing manipulation skills. In this paper, we implement a method that emulates the curiosity drive and uses it as a heuristic to guide (simulated) exploration of a particular task – pouring liquids. The result of this exploration is a collection of symbolic rules linking qualitative descriptions of object arrangements and the pouring action with qualitative descriptions of likely outcomes. The manner in which qualitative descriptions of object arrangements and actions are converted to numerical descriptions for the purpose of simulation parametrization is via probability distributions, which themselves are adjusted in the process of simulated exploration. This allows the grounding of the symbolic descriptions to attempt to adapt itself to the task. The resulting symbolic rules form a theory that, together with the probability distributions that ground it in numerical parametrizations, is intended to be used to predict qualitative outcomes or select manners of pouring towards achieving a goal.

In this paper, we present foundations of the Socio-physical Model of Activities (SOMA). SOMA represents both the physical as well as the social context of everyday activities. Such tasks seem to be trivial for humans, however, they pose severe problems for artificial agents. For starters, a natural language command requesting something will leave many pieces of information necessary for performing the task unspecified. Humans can solve such problems fast as we reduce the search space by recourse to prior knowledge such as a connected collection of plans that describe how certain goals can be achieved at various levels of abstraction. Rather than enumerating fine-grained physical contexts SOMA sets out to include socially constructed knowledge about the functions of actions to achieve a variety of goals or the roles objects can play in a given situation. As the human cognition system is capable of generalizing experiences into abstract knowledge pieces applicable to novel situations, we argue that both physical and social context need be modeled to tackle these challenges in a general manner. The central contribution of this work, therefore, lies in a comprehensive model connecting physical and social entities, that enables flexibility of executions by the robotic agents via symbolic reasoning with the model. This is, by and large, facilitated by the link between the physical and social context in SOMA where relationships are established between occurrences and generalizations of them, which has been demonstrated in several use cases in the domain of everyday activites that validate SOMA.

Functional relations such as containment or support have proven difficult to formalize. Although previous efforts have attempted this using hybrids of several theories, from mereology to temporal logic, we find that such purely symbolic approaches do not account for the embodied nature of functional relations, i.e. that they are used by embodied agents to describe fragments of a physical world. We propose a formalism that combines descriptions of a high level of abstraction with generative models that can be used to instantiate or recognize arrangements of objects and trajectories conforming to qualitative descriptions. The formalism gives an account of how a qualitative description of a scene or arrangement of objects can be converted into a quantitative description amenable to simulation, and how simulation results can be qualitatively interpreted. We use this to describe functional relations between objects in terms of spatial arrangements, expectations on behavior, and counterfactual expectations for when one of the participants is absent. Our method is able to tackle important questions facing an agent operating in the world, such as what would happen if an arrangement of objects is created and why. This gives the agent a deeper understanding of functional relations, including what role background objects, not explicitly asserted to participate in a functional relation such as containment, play in enabling or hindering the relation from holding.

In this paper we present KnowRob2, a second generation knowledge representation and reasoning framework for robotic agents. KnowRob2 is an extension and partial redesign of KnowRob, currently one of the most advanced knowledge processing systems for robots that has enabled them to successfully perform complex manipulation tasks such as making pizza, conducting chemical experiments, and setting tables. The knowledge base appears to be a conventional first-order time interval logic knowledge base, but it exists to a large part only virtually: many logical expressions are constructed on demand from data structures of the control program, computed through robotics algorithms including ones for motion planning and solving inverse kinematics problems, and log data stored in noSQL databases. Novel features and extensions of KnowRob2 substantially increase the capabilities of robotic agents of acquiring open-ended manipulation skills and competence, reasoning about how to perform manipulation actions more realistically, and acquiring commonsense knowledge.

With the advancements in robotic technology and the progress in human-robot interaction research, the interest in deploying mixed human-robot teams in rescue missions is increasing. Due to their complementary capabilities in terms of locomotion, visibility and reachability of areas, human-robot teams are considerably deployed in real-world settings, albeit the robotic agents in such scenarios are normally fully teleoperated. A major barrier to successful and efficient mission execution in those teams is the lack of cognitive skills in robotic systems. In this paper, we present a cognition-enabled framework and an implemented system where robotic agents are equipped with cognitive capabilities to naturally communicate with humans and autonomously perform tasks. The framework allows for natural tasking of robots, reasoning about robot behavior, capabilities and actions, and a common belief state representation for shared mission awareness of robots and human operators.

One intuitive approach for planning robotic manipulation tasks is to first compute a plan for the manipulated object, as if capable to move on its own, then use the obtained plan to find a sequence of arm maneuvers to take the object along the computed plan. Motion planning queries would be performed in relatively low-dimensional configuration spaces, rather than the full configuration space of a robot's arms and grippers, resulting in a more efficient planning method. However, having a plan for the manipulated object does not guarantee there also exists a feasible sequence of maneuvers for the robot to do the manipulation. Knowing different possible grasps on the object increases the robots chance to find a sequence of grasp switches and maneuvers, but makes the search more time consuming as the grasps need to be tested for usefulness. In this paper, we develop a multi-level architecture for complex manipulation planning of rigid bodies which uses communication between the two levels, one for planning the manipulated object motion, the other to plan for the arms, to improve the efficiency of the method. We store grasp zones in the configuration space of the manipulated object as regions where a given grasp seems promising. We use grasp zones to guide searching for grasp switching maneuvers, and to avoid regions of configuration space where few good grasps exist.

This paper introduces an ultra-tightly coupled approach to data fusion of optical and inertial measurements. The two redundant sensor systems complement each other well, with the cameras providing absolute positions and the inertial measurements giving low latency information of derivatives. The targeted application is the tracking of handheld input devices for robotic surgery, where landmarks are not always visible to all cameras. Especially when bi-manual operation is considered, where one hand can move between the other hand and a camera, occlusions occur frequently. The ultra-tighly coupled data fusion uses 2D-camera measurements to correct pose estimations in an extended Kalman filter without an explicit 3D-reconstruction. Therefore marker measurements are used to support the pose estimation, even if the marker is only visible in one camera. Experiments were done with an inertial measurement unit and rectified stereo cameras that show the advantage of the approach for the application.

This paper proposes a method for accurate robotic motion compensation of a freely moving target object. This approaches a typical problem in medical scenarios, where a robotic system needs to compensate physiological movements of a target region related to the patient. An optical tracking system measures the poses of the robot's end-effector and the moving target. The task is to track the target with the robot in a desired relative pose. Arbitrary motion in 6 DoF is covered. The position controller of the medical light-weight robot MIRO is enhanced by a Cartesian displacement observer. The proposed observer feedback preserves the dynamics of the robot, while achieving high accuracy in task space. The target object is equipped with an inertial measurement unit in addition to tracking markers. Target sensor data is fused by an extended Kalman filter in a tightly coupled approach. The robot control and the target tracking in the task space aim to combine accuracy, dynamic performance and robustness to marker occlusions. The algorithms are verified with the DLR MIRO, an experimental target platform, and a commercial tracking system. The experiments demonstrate rapid convergence to desired Cartesian poses and good dynamic tracking performance even at higher target motion speed.

In this paper a sensor fusion for pose estimation using optical and inertial data is presented. The proposed algorithm is based on extended Kalman filtering and fuses data from an optical tracking system and an inertial measurement unit. These two redundant sensor systems complement each other well, with the tracking system providing absolute positions and the inertial measurements giving low latency information of derivatives. Models for both sensors are given respecting the different sampling times and latencies. Another key issue is to use information about every landmark, i.e. marker ball, visible for the tracking system, by coupling the two sensor systems tightly together. The algorithm is evaluated in simulation and tested with an experimental hardware platform. The combined sensor system is robust with respect to short time marker occlusions and effectively compensates for latencies in the pose measurements.