Aude Billard

Noémie Jaquier, Michael C. Welle, Andrej Gams et al.

Transfer learning is a conceptually-enticing paradigm in pursuit of truly intelligent embodied agents. The core concept -- reusing prior knowledge to learn in and from novel situations -- is successfully leveraged by humans to handle novel situations. In recent years, transfer learning has received renewed interest from the community from different perspectives, including imitation learning, domain adaptation, and transfer of experience from simulation to the real world, among others. In this paper, we unify the concept of transfer learning in robotics and provide the first taxonomy of its kind considering the key concepts of robot, task, and environment. Through a review of the promises and challenges in the field, we identify the need of transferring at different abstraction levels, the need of quantifying the transfer gap and the quality of transfer, as well as the dangers of negative transfer. Via this position paper, we hope to channel the effort of the community towards the most significant roadblocks to realize the full potential of transfer learning in robotics.

Diego Paez-Granados, Yujie He, David Gonon et al.

Autonomous navigation in highly populated areas remains a challenging task for robots because of the difficulty in guaranteeing safe interactions with pedestrians in unstructured situations. In this work, we present a crowd navigation control framework that delivers continuous obstacle avoidance and post-contact control evaluated on an autonomous personal mobility vehicle. We propose evaluation metrics for accounting efficiency, controller response and crowd interactions in natural crowds. We report the results of over 110 trials in different crowd types: sparse, flows, and mixed traffic, with low- (< 0.15 ppsm), mid- (< 0.65 ppsm), and high- (< 1 ppsm) pedestrian densities. We present comparative results between two low-level obstacle avoidance methods and a baseline of shared control. Results show a 10% drop in relative time to goal on the highest density tests, and no other efficiency metric decrease. Moreover, autonomous navigation showed to be comparable to shared-control navigation with a lower relative jerk and significantly higher fluency in commands indicating high compatibility with the crowd. We conclude that the reactive controller fulfils a necessary task of fast and continuous adaptation to crowd navigation, and it should be coupled with high-level planners for environmental and situational awareness.

Bernardo Fichera, Viacheslav Borovitskiy, Andreas Krause et al.

Gaussian process regression is widely used because of its ability to provide well-calibrated uncertainty estimates and handle small or sparse datasets. However, it struggles with high-dimensional data. One possible way to scale this technique to higher dimensions is to leverage the implicit low-dimensional manifold upon which the data actually lies, as postulated by the manifold hypothesis. Prior work ordinarily requires the manifold structure to be explicitly provided though, i.e. given by a mesh or be known to be one of the well-known manifolds like the sphere. In contrast, in this paper we propose a Gaussian process regression technique capable of inferring implicit structure directly from data (labeled and unlabeled) in a fully differentiable way. For the resulting model, we discuss its convergence to the Matérn Gaussian process on the assumed manifold. Our technique scales up to hundreds of thousands of data points, and may improve the predictive performance and calibration of the standard Gaussian process regression in high-dimensional settings.

Baiyu Peng, Aude Billard

Planning for diverse real-world robotic tasks necessitates to know and write all constraints. However, instances exist where these constraints are either unknown or challenging to specify accurately. A possible solution is to infer the unknown constraints from expert demonstration. This paper presents a novel two-step Positive-Unlabeled Constraint Learning (PUCL) algorithm to infer a continuous constraint function from demonstrations, without requiring prior knowledge of the true constraint parameterization or environmental model as existing works. We treat all data in demonstrations as positive (feasible) data, and learn a control policy to generate potentially infeasible trajectories, which serve as unlabeled data. The proposed two-step learning framework first identifies reliable infeasible data using a distance metric, and secondly learns a binary feasibility classifier (i.e., constraint function) from the feasible demonstrations and reliable infeasible data. The proposed method is flexible to learn complex-shaped constraint boundary and will not mistakenly classify demonstrations as infeasible as previous methods. The effectiveness of the proposed method is verified in four constrained environments, using a networked policy or a dynamical system policy. It successfully infers the continuous nonlinear constraints and outperforms other baseline methods in terms of constraint accuracy and policy safety. This work has been published in IEEE Robotics and Automation Letters (RA-L). Please refer to the final version at https://doi.org/10.1109/LRA.2024.3522756

Baiyu Peng, Aude Billard

Planning for a wide range of real-world tasks necessitates to know and write all constraints. However, instances exist where these constraints are either unknown or challenging to specify accurately. A possible solution is to infer the unknown constraints from expert demonstration. The majority of prior works limit themselves to learning simple linear constraints, or require strong knowledge of the true constraint parameterization or environmental model. To mitigate these problems, this paper presents a positive-unlabeled (PU) learning approach to infer a continuous, arbitrary and possibly nonlinear, constraint from demonstration. From a PU learning view, We treat all data in demonstrations as positive (feasible) data, and learn a (sub)-optimal policy to generate high-reward-winning but potentially infeasible trajectories, which serve as unlabeled data containing both feasible and infeasible states. Under an assumption on data distribution, a feasible-infeasible classifier (i.e., constraint model) is learned from the two datasets through a postprocessing PU learning technique. The entire method employs an iterative framework alternating between updating the policy, which generates and selects higher-reward policies, and updating the constraint model. Additionally, a memory buffer is introduced to record and reuse samples from previous iterations to prevent forgetting. The effectiveness of the proposed method is validated in two Mujoco environments, successfully inferring continuous nonlinear constraints and outperforming a baseline method in terms of constraint accuracy and policy safety.

Mario Andres Chavarria, Santiago Price Torrendell, Aude Billard et al.

Robotic wheelchairs (RWs) offer significant potential to enhance autonomy and participation for people with mobility impairments, yet many systems have failed to achieve sustained real-world adoption. This narrative literature review examined the extent and quality of end-user involvement in RW design, development, and evaluation over the past decade (2015--2025), assessed against core principles shared by major user-involvement approaches (e.g., user-/human-centered design, participatory/co-design, and inclusive design). The findings indicate that user involvement remains limited and is predominantly concentrated in late-stage evaluation rather than in early requirements definition or iterative co-design. Of the 399 records screened, only 23 studies (about 6%) met the inclusion criteria of verifiable end-user involvement, and many relied on small samples, often around ten participants, with limited justification for sample size selection, proxy users, laboratory-based validation, and non-standardized feedback methods. Research teams were largely engineering-dominated (about 89%) and geographically concentrated in high-income countries. Despite strong evidence that sustained user engagement improves usability and adoption in assistive technology, its systematic implementation in RW research remains rare. Advancing the field requires embedding participatory methodologies throughout the design lifecycle and addressing systemic barriers that constrain meaningful user involvement.

Ahmet Tekden, Dimitrios Kanoulas, Aude Billard et al.

Dynamical systems (DS) methods for Learning-from-Demonstration (LfD) provide stable, continuous policies from few demonstrations. First-order dynamical systems (DS) are effective for many point-to-point and periodic tasks, as long as a unique velocity is defined for each state. For tasks with intersections (e.g., drawing an "8"), extensions such as second-order dynamics or phase variables are often used. However, by incorporating velocity, second-order models become sensitive to disturbances near intersections, as velocity is used to disambiguate motion direction. Moreover, this disambiguation may fail when nearly identical position-velocity pairs correspond to different onward motions. In contrast, phase-based methods rely on open-loop time or phase variables, which limit their ability to recover after perturbations. We introduce Phase-varying Neural Potential Functions (PNPF), an LfD framework that conditions a potential function on a phase variable which is estimated directly from state progression, rather than on open-loop temporal inputs. This phase variable allows the system to handle state revisits, while the learned potential function generates local vector fields for reactive and stable control. PNPF generalizes effectively across point-to-point, periodic, and full 6D motion tasks, outperforms existing baselines on trajectories with intersections, and demonstrates robust performance in real-time robotic manipulation under external disturbances.

Bernardo Fichera, Aude Billard

Dynamical Systems (DS) are an effective and powerful means of shaping high-level policies for robotics control. They provide robust and reactive control while ensuring the stability of the driving vector field. The increasing complexity of real-world scenarios necessitates DS with a higher degree of non-linearity, along with the ability to adapt to potential changes in environmental conditions, such as obstacles. Current learning strategies for DSs often involve a trade-off, sacrificing either stability guarantees or offline computational efficiency in order to enhance the capabilities of the learned DS. Online local adaptation to environmental changes is either not taken into consideration or treated as a separate problem. In this paper, our objective is to introduce a method that enhances the complexity of the learned DS without compromising efficiency during training or stability guarantees. Furthermore, we aim to provide a unified approach for seamlessly integrating the initially learned DS's non-linearity with any local non-linearities that may arise due to changes in the environment. We propose a geometrical approach to learn asymptotically stable non-linear DS for robotics control. Each DS is modeled as a harmonic damped oscillator on a latent manifold. By learning the manifold's Euclidean embedded representation, our approach encodes the non-linearity of the DS within the curvature of the space. Having an explicit embedded representation of the manifold allows us to showcase obstacle avoidance by directly inducing local deformations of the space. We demonstrate the effectiveness of our methodology through two scenarios: first, the 2D learning of synthetic vector fields, and second, the learning of 3D robotic end-effector motions in real-world settings.

Aude Billard, Alin Albu-Schaeffer, Michael Beetz et al.

AI technologies, including deep learning, large-language models have gone from one breakthrough to the other. As a result, we are witnessing growing excitement in robotics at the prospect of leveraging the potential of AI to tackle some of the outstanding barriers to the full deployment of robots in our daily lives. However, action and sensing in the physical world pose greater and different challenges than analysing data in isolation. As the development and application of AI in robotic products advances, it is important to reflect on which technologies, among the vast array of network architectures and learning models now available in the AI field, are most likely to be successfully applied to robots; how they can be adapted to specific robot designs, tasks, environments; which challenges must be overcome. This article offers an assessment of what AI for robotics has achieved since the 1990s and proposes a short- and medium-term research roadmap listing challenges and promises. These range from keeping up-to-date large datasets, representatives of a diversity of tasks robots may have to perform, and of environments they may encounter, to designing AI algorithms tailored specifically to robotics problems but generic enough to apply to a wide range of applications and transfer easily to a variety of robotic platforms. For robots to collaborate effectively with humans, they must predict human behavior without relying on bias-based profiling. Explainability and transparency in AI-driven robot control are not optional but essential for building trust, preventing misuse, and attributing responsibility in accidents. We close on what we view as the primary long-term challenges, that is, to design robots capable of lifelong learning, while guaranteeing safe deployment and usage, and sustainable computational costs.

Luyin Hu, Soheil Gholami, George Dindelegan et al.

Microsurgical anastomosis demands exceptional dexterity and visuospatial skills, underscoring the importance of comprehensive training and precise outcome assessment. Currently, methods such as the outcome-oriented anastomosis lapse index are used to evaluate this procedure. However, they often rely on subjective judgment, which can introduce biases that affect the reliability and efficiency of the assessment of competence. Leveraging three datasets from hospitals with participants at various levels, we introduce a quantitative framework that uses image-processing techniques for objective assessment of microsurgical anastomoses. The approach uses geometric modeling of errors along with a detection and scoring mechanism, enhancing the efficiency and reliability of microsurgical proficiency assessment and advancing training protocols. The results show that the geometric metrics effectively replicate expert raters' scoring for the errors considered in this work.

Yang Liu, Bruno Da Costa, Aude Billard

Dynamic manipulation, such as robot tossing or throwing objects, has recently gained attention as a novel paradigm to speed up logistic operations. However, the focus has predominantly been on the object's landing location, irrespective of its final orientation. In this work, we present a method enabling a robot to accurately "throw-flip" objects to a desired landing pose (position and orientation). Conventionally, objects thrown by revolute robots suffer from parasitic rotation, resulting in highly restricted and uncontrollable landing poses. Our approach is based on two key design choices: first, leveraging the impulse-momentum principle, we design a family of throwing motions that effectively decouple the parasitic rotation, significantly expanding the feasible set of landing poses. Second, we combine a physics-based model of free flight with regression-based learning methods to account for unmodeled effects. Real robot experiments demonstrate that our framework can learn to throw-flip objects to a pose target within ($\pm$5 cm, $\pm$45 degrees) threshold in dozens of trials. Thanks to data assimilation, incorporating projectile dynamics reduces sample complexity by an average of 40% when throw-flipping to unseen poses compared to end-to-end learning methods. Additionally, we show that past knowledge on in-hand object spinning can be effectively reused, accelerating learning by 70% when throwing a new object with a Center of Mass (CoM) shift. A video summarizing the proposed method and the hardware experiments is available at https://youtu.be/txYc9b1oflU.

Bernardo Fichera, Aude Billard

Dynamical Systems (DS) are fundamental to the modeling and understanding time evolving phenomena, and have application in physics, biology and control. As determining an analytical description of the dynamics is often difficult, data-driven approaches are preferred for identifying and controlling nonlinear DS with multiple equilibrium points. Identification of such DS has been treated largely as a supervised learning problem. Instead, we focus on an unsupervised learning scenario where we know neither the number nor the type of dynamics. We propose a Graph-based spectral clustering method that takes advantage of a velocity-augmented kernel to connect data points belonging to the same dynamics, while preserving the natural temporal evolution. We study the eigenvectors and eigenvalues of the Graph Laplacian and show that they form a set of orthogonal embedding spaces, one for each sub-dynamics. We prove that there always exist a set of 2-dimensional embedding spaces in which the sub-dynamics are linear and n-dimensional embedding spaces where they are quasi-linear. We compare the clustering performance of our algorithm to Kernel K-Means, Spectral Clustering and Gaussian Mixtures and show that, even when these algorithms are provided with the correct number of sub-dynamics, they fail to cluster them correctly. We learn a diffeomorphism from the Laplacian embedding space to the original space and show that the Laplacian embedding leads to good reconstruction accuracy and a faster training time through an exponential decaying loss compared to the state-of-the-art diffeomorphism-based approaches.

Ahalya Prabhakar, Stanislas Furrer, Lorenzo Panchetti et al.

Adaptive control for real-time manipulation requires quick estimation and prediction of object properties. While robot learning in this area primarily focuses on using vision, many tasks cannot rely on vision due to object occlusion. Here, we formulate a learning framework that uses multimodal sensory fusion of tactile and audio data in order to quickly characterize and predict an object's properties. The predictions are used in a developed reactive controller to adapt the grip on the object to compensate for the predicted inertial forces experienced during motion. Drawing inspiration from how humans interact with objects, we propose an experimental setup from which we can understand how to best utilize different sensory signals and actively interact with and manipulate objects to quickly learn their object properties for safe manipulation.

Ahalya Prabhakar, Aude Billard

A critical need in assistive robotics, such as assistive wheelchairs for navigation, is a need to learn task intent and safety guarantees through user interactions in order to ensure safe task performance. For tasks where the objectives from the user are not easily defined, learning from user demonstrations has been a key step in enabling learning. However, most robot learning from demonstration (LfD) methods primarily rely on optimal demonstration in order to successfully learn a control policy, which can be challenging to acquire from novice users. Recent work does use suboptimal and failed demonstrations to learn about task intent; few focus on learning safety guarantees to prevent repeat failures experienced, essential for assistive robots. Furthermore, interactive human-robot learning aims to minimize effort from the human user to facilitate deployment in the real-world. As such, requiring users to label the unsafe states or keyframes from the demonstrations should not be a necessary requirement for learning. Here, we propose an algorithm to learn a safety value function from a set of suboptimal and failed demonstrations that is used to generate a real-time safety control filter. Importantly, we develop a credit assignment method that extracts the failure states from the failed demonstrations without requiring human labelling or prespecified knowledge of unsafe regions. Furthermore, we extend our formulation to allow for user-specific safety functions, by incorporating user-defined safety rankings from which we can generate safety level sets according to the users' preferences. By using both suboptimal and failed demonstrations and the developed credit assignment formulation, we enable learning a safety value function with minimal effort needed from the user, making it more feasible for widespread use in human-robot interactive learning tasks.

Lukas Huber, Jean-Jacques Slotine, Aude Billard

This paper presents a closed-form approach to constrain a flow within a given volume and around objects. The flow is guaranteed to converge and to stop at a single fixed point. We show that the obstacle avoidance problem can be inverted to enforce that the flow remains enclosed within a volume defined by a polygonal surface. We formally guarantee that such a flow will never contact the boundaries of the enclosing volume and obstacles, and will asymptotically converge towards an attractor. We further create smooth motion fields around obstacles with edges (e.g. tables). Both obstacles and enclosures may be time-varying, i.e. moving, expanding and shrinking. The technique enables a robot to navigate within an enclosed corridor while avoiding static and moving obstacles. It was applied on an autonomous robot (QOLO) in a static complex indoor environment, and also tested in simulations with dense crowds. The final proof of concept was performed in an outdoor environment in Lausanne. The QOLO-robot successfully traversed a marketplace in the center of town in presence of a diverse crowd with a non-uniform motion pattern.

Fabien Grzeskowiak, David Gonon, Daniel Dugas et al.

The evaluation of robot capabilities to navigate human crowds is essential to conceive new robots intended to operate in public spaces. This paper initiates the development of a benchmark tool to evaluate such capabilities; our long term vision is to provide the community with a simulation tool that generates virtual crowded environment to test robots, to establish standard scenarios and metrics to evaluate navigation techniques in terms of safety and efficiency, and thus, to install new methods to benchmarking robots' crowd navigation capabilities. This paper presents the architecture of the simulation tools, introduces first scenarios and evaluation metrics, as well as early results to demonstrate that our solution is relevant to be used as a benchmark tool.

Jacob Hernandez Sanchez, Walid Amanhoud, Anaïs Haget et al.

We seek to augment human manipulation by enabling humans to control two robotic arms in addition to their natural arms using their feet. Thereby, the hands are free to perform tasks of high dexterity, while the feet-controlled arms perform tasks requiring lower dexterity, such as supporting a load. The robotic arms are tele-operated through two foot interfaces that transmit translation and rotation to the end effector of the manipulator. Haptic feedback is provided for the human to perceive contact and change in load and to adapt the feet pressure accordingly. Existing foot interfaces have been used primarily for a single foot control and are limited in range of motion and number of degrees of freedom they can control. This paper presents foot-interfaces specifically made for bipedal control, with a workspace suitable for two feet operation and in five degrees of freedom each. This paper also presents a position-force teleoperation controller based on Impedance Control modulated through Dynamical Systems for trajectory generation. Finally, an initial validation of the platform is presented, whereby a user grasps an object with both feet and generates various disturbances while the object is supported by the feet.

Nuno Ferreira Duarte, Jovica Tasevski, Moreno Coco et al.

Humans have the fascinating capacity of processing non-verbal visual cues to understand and anticipate the actions of other humans. This "intention reading" ability is underpinned by shared motor-repertoires and action-models, which we use to interpret the intentions of others as if they were our own. We investigate how the different cues contribute to the legibility of human actions during interpersonal interactions. Our first contribution is a publicly available dataset with recordings of human body-motion and eye-gaze, acquired in an experimental scenario with an actor interacting with three subjects. From these data, we conducted a human study to analyse the importance of the different non-verbal cues for action perception. As our second contribution, we used the motion/gaze recordings to build a computational model describing the interaction between two persons. As a third contribution, we embedded this model in the controller of an iCub humanoid robot and conducted a second human study, in the same scenario with the robot as an actor, to validate the model's "intention reading" capability. Our results show that it is possible to model (non-verbal) signals exchanged by humans during interaction, and how to incorporate such a mechanism in robotic systems with the twin goal of : (i) being able to "read" human action intentions, and (ii) acting in a way that is legible by humans.

Nadia Figueroa, Aude Billard

In this work, we tackle the problem of transform-invariant unsupervised learning in the space of Covariance matrices and applications thereof. We begin by introducing the Spectral Polytope Covariance Matrix (SPCM) Similarity function; a similarity function for Covariance matrices, invariant to any type of transformation. We then derive the SPCM-CRP mixture model, a transform-invariant non-parametric clustering approach for Covariance matrices that leverages the proposed similarity function, spectral embedding and the distance-dependent Chinese Restaurant Process (dd-CRP) (Blei and Frazier, 2011). The scalability and applicability of these two contributions is extensively validated on real-world Covariance matrix datasets from diverse research fields. Finally, we couple the SPCM-CRP mixture model with the Bayesian non-parametric Indian Buffet Process (IBP) - Hidden Markov Model (HMM) (Fox et al., 2009), to jointly segment and discover transform-invariant action primitives from complex sequential data. Resulting in a topic-modeling inspired hierarchical model for unsupervised time-series data analysis which we call ICSC-HMM (IBP Coupled SPCM-CRP Hidden Markov Model). The ICSC-HMM is validated on kinesthetic demonstrations of uni-manual and bi-manual cooking tasks; achieving unsupervised human-level decomposition of complex sequential tasks.

Manuel Wüthrich, Peter Pastor, Ludovic Righetti et al.

In this paper, we derive a probabilistic registration algorithm for object modeling and tracking. In many robotics applications, such as manipulation tasks, nonvisual information about the movement of the object is available, which we will combine with the visual information. Furthermore we do not only consider observations of the object, but we also take space into account which has been observed to not be part of the object. Furthermore we are computing a posterior distribution over the relative alignment and not a point estimate as typically done in for example Iterative Closest Point (ICP). To our knowledge no existing algorithm meets these three conditions and we thus derive a novel registration algorithm in a Bayesian framework. Experimental results suggest that the proposed methods perform favorably in comparison to PCL implementations of feature mapping and ICP, especially if nonvisual information is available.