Initiatives et contributions récentes des Axes

This paper presents significant advancements in the detection and management of human-cable collisions within Cable-Driven Parallel Robots. Building upon previous research [1], A novel frequencybased filter is developed and applied to tension sensor measurements, enabling collision detection based on cable tension measurements. This approach facilitates the identification of the colliding cable and allows for the reduction of the concerned cable tension, permitting safe contact between the cable and the environment or operator without causing damage. Additionally, a robust method for detecting the end of collisions is proposed, ensuring the system can promptly return to normal operation. An adaptive control method for cable length release is also developed, optimizing collision management during dynamic human-robot interactions. The proposed management approach effectively handles both severe and minor collisions. Experiments conducted with the CRAFT prototype validate these improvements, demonstrating that they substantially enhance safety and responsiveness in physical human-robot collaboration, thereby marking a noteworthy progression in collaborative robotic environments.

Robotic designs played an important role in recent advances by providing powerful robots with complex mechanics. Many recent systems rely on parallel actuation to provide lighter limbs and allow more complex motion. However, these emerging architectures fall outside the scope of most used description formats, leading to difficulties when designing, storing, and sharing the models of these systems. This paper introduces an extension to the widely used Unified Robot Description Format (URDF) to support closed-loop kinematic structures. Our approach relies on augmenting URDF with minimal additional information to allow more efficient modeling of complex robotic systems while maintaining compatibility with existing design and simulation frameworks. This method sets the basic requirement for a description format to handle parallel mechanisms efficiently. We demonstrate the applicability of our approach by providing an open-source collection of parallel robots, along with tools for generating and parsing this extended description format. The proposed extension simplifies robot modeling, reduces redundancy, and improves usability for advanced robotic applications.

Despite its various applications, robotic manipulation of deformable objects in agriculture has experienced limited development so far. This is due to the specific challenges in this domain, i.e., the variety of objects in this field is wide, and the deformation properties of the objects cannot be easily recognized in advance. In addition, deformable objects generally have complex dynamics and high-dimensional configuration space. In this paper, the manipulation of deformable linear objects (DLOs) is addressed by considering these challenges. Concretely, a new indirect adaptive control method is proposed to manipulate DLOs by controlling their shape in 3-D space towards previously defined targets, with a specific focus on agricultural applications. The proposed method can follow a desired dynamic evolution of the shape with a smooth deformation that brings about a stable gripper motion. This property of the method can protect the object from possible damages, even under large deformations, which is crucial in agricultural scenarios. An adaptation law is leveraged for estimating the system parameters, and Lyapunov analysis is employed to study the validity of the proposed control scheme. The scheme can be applied to diverse objects that can be modeled as linear, including tree branches or other rod-like structures. The effectiveness of the scheme is demonstrated through various experiments where, using shape feedback obtained from a 3-D camera, a robotic arm controls the shape of a flexible foam rod and of branches of different plants. © 2025

While silicone printing by extrusion is a promising technique, part production is still a challenge when supportfree printing is considered. In this paper, we consider three main causes of defects, namely the management of silicone flow rate during printing, the part deformation under its own weight before polymerization and the deformations due to the nozzle-layer interactions during printing. An instrumentation strategy is here proposed to monitor layer deposition. Our approach, based on laser profilometry, is detailed with open source software to control printing and scanning. The process to build point clouds and extract data is presented. More importantly, indicators are introduced to build metrics related to the current main causes of printing failure, using prior geometric information on the planned layers. Through experimental evaluation, the adequacy of the indicators and their complementarity is discussed. This introduction of new indicators opens ways to implement efficient silicone extrusion supervision.

Ultrasound (US) imaging enables the evaluation of vascular structures in real time, and it can provide morphological and pathological information during US-guided procedures. Automatic prediction of vascular structure boundaries can help clinicians in locating and measuring target structures more accurately and efficiently. Most existing US segmentation methods use per-pixel classification or regression, which require post-processing to obtain contour coordinates. In this work, we present ACE-Net, a novel approach that directly predicts the contour coordinates for every scanning line (A-line) in US images. ACE-Net combines two main modules: a boundary regression module that predicts the upper and lower coordinates of the target area for each A-line, and an A-line classification module that determines whether an A-line belongs to the target area or not. We evaluated our method on three clinical US datasets using, among others, dice similarity coefficient (DSC) and inference time as performance metrics. Our method outperformed state-of-the-art segmentation methods in inference time while achieving superior or comparable performance in DSC. ACE-Net is publicly available at https://github.com/bfarolabarata/ace-net.