Contact-centric tasks on surfaces, ranging from inspection and cleaning to sanding and polishing, require robots to systematically cover the surface while maintaining stable contact. Ergodic control generates trajectories that spend time at a location proportional to a desired, task-specific spatial distribution, enabling efficient information gathering and coverage. However, traditional ergodic control methods rely on prior knowledge of surface geometry or require a vision sensory input to scan the geometry beforehand, limiting their applicability in real-world scenarios with unknown or dynamic environments. This paper introduces a novel online ergodic control framework that achieves systematic surface coverage while simultaneously reconstructing unknown surface geometry. We employ a Gaussian Process Implicit Surface (GPIS) model that learns global surface geometry from intrinsic tactile sensing during execution. For efficient online planning, we approximate the surface locally using point clouds sampled from tangent planes at observed contact points and iteratively fit them to the Gaussian Process. This approximation simultaneously serves as the sampling domain for both the target and the coverage distributions. We employ a heat-diffusion analogy to compute potential fields that guide ergodic exploration, translating spatial coverage objectives into smooth robot trajectories. We demonstrate our framework through simulation and real-robot experiments, validating simultaneous ergodic coverage and online surface geometry learning with reconstruction errors approaching the ground truth.

Industrial robot applications require increasingly flexible systems that non-expert users can easily adapt for varying tasks and environments. However, different adaptations benefit from different interaction modalities. We present an interactive framework that enables robot skill adaptation through three complementary modalities: kinesthetic touch for precise spatial corrections, natural language for high-level semantic modifications, and a graphical web interface for visualizing geometric relations and trajectories, inspecting and adjusting parameters, and editing via-points by drag-and-drop. The framework integrates five components: energy-based human-intention detection, a tool-based LLM architecture (where the LLM selects and parameterizes predefined functions rather than generating code) for safe natural language adaptation, Kernelized Movement Primitives (KMPs) for motion encoding, probabilistic Virtual Fixtures for guided demonstration recording, and ergodic control for surface finishing. We demonstrate that this tool-based LLM architecture generalizes skill adaptation from KMPs to ergodic control, enabling voice-commanded surface finishing. Validation on a 7-DoF torque-controlled robot at the Automatica 2025 trade fair demonstrates the practical applicability of our approach in industrial settings.

Robots are crucial for exploring distant celestial bodies. The Surface Avatar ISS-to-Earth Telerobotic Technology Demonstration Mission investigates how to command a heterogeneous team of robots from orbit using scalable autonomy. In this experiment series astronauts aboard the International Space Station (ISS) command a team of robots located on Earth. One challenge identified in these experiments is that the robot’s decision-making process is affected by environmental uncertainty especially for the robot’s pose estimation and navigation. Traditional planning algorithms assume perfect knowledge of the robot’s surroundings, neglecting the role of imperfect sensing. This paper addresses this limitation by developing an uncertainty-aware planning method and demonstrating its application to address perception inaccuracies in navigation, paving the way for its broader adoption in other domains. The proposed uncertainty aware planning framework is tested for localization and navigation by the humanoid robot Rollin’ Justin during a space-to-ground telerobotic experiment as part of the Surface Avatar mission.

We introduce NealAI, the first AI chat assistant to support astronauts with question answering during a space telerobotics experiment. In the Surface Avatar mission, an ISS crew member controlled a heterogeneous team of four robots in a simulated Martian environment. NealAI uses a Retrieval-Augmented Generation (RAG) approach, enabling a Large Language Model (LLM) to dynamically retrieve relevant context about the experiment and its robots, and deliver accurate, context-aware responses. To adhere to privacy requirements and computational costs, NealAI is based on a single small-scale LLM running locally. We assessed NealAI’s performance in different evaluations, including a preliminary experiment with an ISS crew member teleoperating the robots, as well as a set of offline tests to evaluate the LLM context selection, the response correctness, and when (and why) hallucinations occur. Results demonstrate the feasibility and limitations of using a small-scale LLM on a RAG-based chat assistant during a space telerobotic experiment. Finally, we report some conclusions and lessons learned.

Future space applications require the assembly of large structures in orbit. This can only be achieved by using autonomous robotic systems able to handle repetitive tasks with heavy and large parts in such challenging environment. This paper provides an overview of a system designed to perform autonomous assembly of segmented mirror tiles, as a proof of feasibility for assembling large structures in space using robotic technologies. We describe the hardware components of the system, and present the software layer, including assembly planning, and skill engine. An experimental evaluation of the assembly process is carried out, thus showing the performance achieved with the system.

This paper presents an automated framework for designing and reconfiguring robotic workcells in high-mix, lowvolume manufacturing scenarios. The framework utilizes a modular workcell with various tools and a flexible grid system, the Variable Workstation (VWS), and uses an ontological representation of available robotic systems and assembly tasks. The major contribution is a tool for automatic layout planning and reconfiguration, and its integration into the overall pipeline. The tool optimizes component placements for efficient task execution with minimal path lengths and maximized manipulability. Constraints guarantee the reachability of each task and absence of collisions. The framework is evaluated in a real scenario using tasks from the assembly of an electric chainsaw. Overall, the proposed framework provides an efficient and complete system of designing and reconfiguring robotic workcells for high-mix, low-volume manufacturing.

The initial design of a robotic workcell layout has a large impact on the feasibility and performance of the intended robotic tasks. We define this layout design as a constrained nonlinear optimization problem that aims to optimize the placement of workcell components by minimizing the distance traveled between task sequences while maximizing the robot’s manipulability. Suitable constraints guarantee the reachability as well as the absence of collisions. We solve this optimization problem via a genetic algorithm, and demonstrate it in three scenarios for a dual-arm robotic system that assembles product variants out of aluminum profiles.

Technology has sufficiently matured to enable, in principle, flexible and autonomous robotic assembly systems. However, in practice, it requires making all the relevant (implicit) knowledge that system engineers and workers have – about products to be assembled, tasks to be performed, as well as robots and their skills – available to the system explicitly. Only then can the planning and execution components of a robotic assembly pipeline communicate with each other in the same language and solve tasks autonomously without human intervention. This is why we have developed the Factory of the Future (FoF) ontology. At its core, this ontology models the tasks that are necessary to assemble a product and the robotic skills that can be employed to complete said tasks. The FoF ontology is based on existing standards. We started with theoretical considerations and iteratively adapted it based on practical experience gained from incorporating more and more components required for automated planning and assembly. Furthermore, we propose tools to extend the ontology for specific scenarios with knowledge about parts, robots, tools, and skills from various sources. The resulting scenario ontology serves us as world model for the robotic systems and other components of the assembly process. A central runtime interface to this world model provides fast and easy access to the knowledge during execution. In this work, we also show the integration of a graphical user front-end, an assembly planner, a workspace reconfigurator, and more components of the assembly pipeline that all communicate with the help of the FoF ontology. Overall, our integration of the FoF ontology with the other components of a robotic assembly pipeline shows that using an ontology is a practical method to establish a common language and understanding between the involved components.

Planning new complex robotic applications requires the knowledge whether and where a specific task is executable within the reachable space of a robotic system. The complexity of a reachability analysis increases when the setup is reconfigurable or consists of multiple robots sharing the same workspace. In this work we show an approach to decouple task-specifications from system-wide configurations to compute task-specific workspace maps. These maps are a powerful tool for designing and planning robotic tasks, since they allow to visualize and suggest feasible poses for task execution. Furthermore, the maps can be queried during runtime to select appropriate poses automatically. Specifying a task manually takes time and effort since it is necessary to define a set of transformations with respect to a task reference frame. To overcome this issue, we show a solution by recording the motion of a robot to generate a task specification. The presented approach offers solutions for flexible systems with reusable components to make robust predictions regarding the kinematic reachability of the robotic setup.