Stefano Stramigioli

Matei Drilea, Alexander Dijkshoorn, Gusthavo Ribeiro Salomão et al.

The excellent structural and piezoresistive properties of continuous carbon fiber make it suitable for both structural and sensing applications. This work studies the use of 3D printed, continuous carbon fiber reinforced beams as self-sensing structures. It is demonstrated how the sensitivity of these carbon fiber strain gauges can be increased irreversibly by means of a pretreatment by pre-stressing the sensors with a large compressive bending load. The increase in the gauge factor is attributed to local progressive fiber failure, due to the combination of the thermal residual stress from the printing process and external loading. The coextrusion of conductive filament around the carbon fibers is demonstrated as a means of improving the reliability, noise and electrical connection of the sensors. A micrograph of the sensor cross section shows that the conductive filament contacts the various carbon fiber bundles. All-in-all, the use of pre-stressing carbon fiber strain gauges in combination with coextrusion of conductive filament hold promises for 3D printed structural sensors with a high sensitivity.

Riccardo Zanella, Federico Califano, Stefano Stramigioli

This paper aims to provide a clear and rigorous understanding of commonly recognized safety constraints in physical human-robot interaction, particularly regarding ISO/TS 15066. We investigate the derivation of these constraints, critically examine the underlying assumptions, and evaluate their practical implications for system-level safety and performance in industrially relevant scenarios. Key design parameters within safety-critical control architectures are identified, and numerical examples are provided to quantify performance degradation arising from typical approximations and design decisions in manufacturing environments. Within this analysis, the fundamental role of energy in safety assessment is emphasized, providing focused insights into energy-based safety methodologies for collaborative industrial robot systems.

Yannik P. Wotte, Patrick Buchfink, Silke Glas et al.

Lie groups and their actions are ubiquitous in the description of physical systems, and we explore implications in the setting of model order reduction (MOR). We present a novel framework of MOR via Lie groups, called MORLie, in which high-dimensional dynamical systems on manifolds are approximated by low-dimensional dynamical systems on Lie groups. In comparison to other Lie group methods we are able to attack non-equivariant dynamics, which are frequent in practical applications, and we provide new non-intrusive MOR methods based on the presented geometric formulation. We also highlight numerically that MORLie has a lower error bound than the Kolmogorov $N$-width, which limits linear-subspace methods. The method is applied to various examples: 1. MOR of a simplified deforming body modeled by noisy point cloud data following a sheering motion, where MORLie outperforms a naive POD approach in terms of accuracy and dimensionality reduction. 2. Reconstructing liver motion during respiration with data from edge detection in MRI scans, where MORLie reaches performance approaching the state of the art, while reducing the training time from hours on a computing cluster to minutes on a mobile workstation. 3. An analytic example showing that the method of freezing is analytically recovered as a special case, showing the generality of the geometric framework.

Bangyu Lan, Momen Abayazid, Nico Verdonschot et al.

In Total Knee Replacement Arthroplasty (TKA), surgical robotics can provide image-guided navigation to fit implants with high precision. Its tracking approach highly relies on inserting bone pins into the bones tracked by the optical tracking system. This is normally done by invasive, radiative manners (implantable markers and CT scans), which introduce unnecessary trauma and prolong the preparation time for patients. To tackle this issue, ultrasound-based bone tracking could offer an alternative. In this study, we proposed a novel deep learning structure to improve the accuracy of bone tracking by an A-mode ultrasound (US). We first obtained a set of ultrasound dataset from the cadaver experiment, where the ground truth locations of bones were calculated using bone pins. These data were used to train the proposed CasAtt-UNet to predict bone location automatically and robustly. The ground truth bone locations and those locations of US were recorded simultaneously. Therefore, we could label bone peaks in the raw US signals. As a result, our method achieved sub millimeter precision across all eight bone areas with the only exception of one channel in the ankle. This method enables the robust measurement of lower extremity bone positions from 1D raw ultrasound signals. It shows great potential to apply A-mode ultrasound in orthopedic surgery from safe, convenient, and efficient perspectives.

Yannik P. Wotte, Federico Califano, Stefano Stramigioli

This work presents a novel approach for the optimization of dynamic systems on finite-dimensional Lie groups. We rephrase dynamic systems as so-called neural ordinary differential equations (neural ODEs), and formulate the optimization problem on Lie groups. A gradient descent optimization algorithm is presented to tackle the optimization numerically. Our algorithm is scalable, and applicable to any finite dimensional Lie group, including matrix Lie groups. By representing the system at the Lie algebra level, we reduce the computational cost of the gradient computation. In an extensive example, optimal potential energy shaping for control of a rigid body is treated. The optimal control problem is phrased as an optimization of a neural ODE on the Lie group SE(3), and the controller is iteratively optimized. The final controller is validated on a state-regulation task.

Nicolò Botteghi, Luuk Grefte, Mannes Poel et al.

Inspection and maintenance are two crucial aspects of industrial pipeline plants. While robotics has made tremendous progress in the mechanic design of in-pipe inspection robots, the autonomous control of such robots is still a big open challenge due to the high number of actuators and the complex manoeuvres required. To address this problem, we investigate the usage of Deep Reinforcement Learning for achieving autonomous navigation of in-pipe robots in pipeline networks with complex topologies. Moreover, we introduce a hierarchical policy decomposition based on Hierarchical Reinforcement Learning to learn robust high-level navigation skills. We show that the hierarchical structure introduced in the policy is fundamental for solving the navigation task through pipes and necessary for achieving navigation performances superior to human-level control.

Nicolò Botteghi, Khaled Alaa, Mannes Poel et al.

Autonomous robots require high degrees of cognitive and motoric intelligence to come into our everyday life. In non-structured environments and in the presence of uncertainties, such degrees of intelligence are not easy to obtain. Reinforcement learning algorithms have proven to be capable of solving complicated robotics tasks in an end-to-end fashion without any need for hand-crafted features or policies. Especially in the context of robotics, in which the cost of real-world data is usually extremely high, reinforcement learning solutions achieving high sample efficiency are needed. In this paper, we propose a framework combining the learning of a low-dimensional state representation, from high-dimensional observations coming from the robot's raw sensory readings, with the learning of the optimal policy, given the learned state representation. We evaluate our framework in the context of mobile robot navigation in the case of continuous state and action spaces. Moreover, we study the problem of transferring what learned in the simulated virtual environment to the real robot without further retraining using real-world data in the presence of visual and depth distractors, such as lighting changes and moving obstacles.

Marta Lagomarsino, Vincent Groenhuis, Maura Casadio et al.

A biopsy is the only diagnostic procedure for accurate histological confirmation of breast cancer. When sonographic placement is not feasible, a Magnetic Resonance Imaging(MRI)-guided biopsy is often preferred. The lack of real-time imaging information and the deformations of the breast make it challenging to bring the needle precisely towards the tumour detected in pre-interventional Magnetic Resonance (MR) images. The current manual MRI-guided biopsy workflow is inaccurate and would benefit from a technique that allows real-time tracking and localisation of the tumour lesion during needle insertion. This paper proposes a robotic setup and software architecture to assist the radiologist in targeting MR-detected suspicious tumours. The approach benefits from image fusion of preoperative images with intraoperative optical tracking of markers attached to the patient's skin. A hand-mounted biopsy device has been constructed with an actuated needle base to drive the tip toward the desired direction. The steering commands may be provided both by user input and by computer guidance. The workflow is validated through phantom experiments. On average, the suspicious breast lesion is targeted with a radius down to 2.3 mm. The results suggest that robotic systems taking into account breast deformations have the potentials to tackle this clinical challenge.

Nicolò Botteghi, Ruben Obbink, Daan Geijs et al.

Reinforcement Learning has been able to solve many complicated robotics tasks without any need for feature engineering in an end-to-end fashion. However, learning the optimal policy directly from the sensory inputs, i.e the observations, often requires processing and storage of a huge amount of data. In the context of robotics, the cost of data from real robotics hardware is usually very high, thus solutions that achieve high sample-efficiency are needed. We propose a method that aims at learning a mapping from the observations into a lower-dimensional state space. This mapping is learned with unsupervised learning using loss functions shaped to incorporate prior knowledge of the environment and the task. Using the samples from the state space, the optimal policy is quickly and efficiently learned. We test the method on several mobile robot navigation tasks in a simulation environment and also on a real robot.

Nicolò Botteghi, Beril Sirmacek, Khaled A. A. Mustafa et al.

We present a map-less path planning algorithm based on Deep Reinforcement Learning (DRL) for mobile robots navigating in unknown environment that only relies on 40-dimensional raw laser data and odometry information. The planner is trained using a reward function shaped based on the online knowledge of the map of the training environment, obtained using grid-based Rao-Blackwellized particle filter, in an attempt to enhance the obstacle awareness of the agent. The agent is trained in a complex simulated environment and evaluated in two unseen ones. We show that the policy trained using the introduced reward function not only outperforms standard reward functions in terms of convergence speed, by a reduction of 36.9\% of the iteration steps, and reduction of the collision samples, but it also drastically improves the behaviour of the agent in unseen environments, respectively by 23\% in a simpler workspace and by 45\% in a more clustered one. Furthermore, the policy trained in the simulation environment can be directly and successfully transferred to the real robot. A video of our experiments can be found at: https://youtu.be/UEV7W6e6ZqI

Ludo C. Visser, Stefano Stramigioli, Raffaella Carloni

Robust and energy-efficient bipedal locomotion in robotics is still a challenging topic. In order to address issues in this field, we can take inspiration from nature, by studying human locomotion. The Spring-Loaded Inverted Pendulum (SLIP) model has shown to be a good model for this purpose. However, the human musculoskeletal system enables us to actively modulate leg stiffness, for example when walking in rough terrain with irregular and unexpected height variations of the walking surface. This ability of varying leg stiffness is not considered in conventional SLIP-based models, and therefore this paper explores the potential role of active leg stiffness variation in bipedal locomotion. It is shown that the conceptual SLIP model can be iteratively extended to more closely resemble a realistic (i.e., non-ideal) walker, and that feedback control strategies can be designed that reproduce the SLIP behavior in these extended models. We show that these extended models realize a cost of transport comparable to human walking, which indicates that active leg stiffness variation plays an important role in human locomotion that was previously not captured by the SLIP model. The results of this study show that active leg stiffness adaptation is a promising approach for realizing more energy-efficient and robust bipedal walking robots.