Pasquale Ferrentino, Ellen Roels, Joost Brancart, Seppe Terryn, Guy Van Assche, Bram Vanderborght
ABSTRACT
Soft robotics modeling is a research topic that is evolving fast. Many techniques are present in literature, but most of them require analytical models with a lot of equations that are time consuming, hard to resolve, and not so easy to handle. For this reason, the help of a soft mechanics simulator is essential in this field. This article presents a tutorial on how to build a soft-robot model using an open source finite element analysis (FEA) simulator, called SOFA. This software is able to generate a simulation scene from a code written in Python or XML, so it can be used by people with different fields of competence, like mechanical knowledge, knowledge of material properties, and programming skills. As a case study, a Python simulation of a cable-driven soft actuator that makes contact with a rigid object is considered. The basic working principles of SOFA required to make a scene are explained step by step. In particular, this article shows how to simulate the mechanics and animate the bending behavior of the actuator and the importance of knowledge of the constitutive material properties for good modeling of the mechanical system. Furthermore, we will also show how to retrieve and save data from simulation, demonstrating that SOFA can easily adapt to a multidisciplinary subject, such as research in soft robotics, but can also be useful for teaching simulation and programming language principles to engineering students.
ABSTRACT
he accumulation of a sufficient amount of data for training deep neural networks is a major hindrance in the application of deep learning in robotics. Acquiring real-world data requires considerable time and effort, yet it might still not capture the full range of potential environmental variations. The generation of new synthetic data based on existing training data has been enabled with the development of generative adversarial networks (GANs). In this paper, we introduce a training methodology based on GANs that utilizes a recurrent, LSTM-based architecture for intention recognition in robotics. The resulting networks predict the intention of the observed human or robot based on input RGB videos. They are trained in a semi-supervised manner, with the output classification networks predicting one of possible labels for the observed motion, while the recurrent generator networks produce fake RGB videos that are leveraged in the training process. We show that utilization of the generated data during the network training process increases the accuracy and generality of motion classification compared to using only real training data. The proposed method can be applied to a variety of dynamic tasks and different LSTM-based classification networks to supplement real data.
ABSTRACT
Occupational exoskeletons are not yet frequently used at work. Previous research has indicated multiple factors (physiological, implementation-related, work-related, policy, and psycho-social) that can explain this lack of adoption. However, there is a lack of specific requirements related to these themes to improve exoskeletons and their adoption. Therefore, the goal of our research is to formulate concrete requirements for the design and implementation of exoskeletons. We used a design ethnography approach (focus groups, an interview, observations, surveys, and group interviews) with multiple stakeholders (users, designers, and potential future users). Our data suggest that potential future users of exoskeletons believe that existing exoskeletons should be improved to be adopted. Exoskeletons should be more comfortable to wear and flexible enough to support a wider range of tasks. We have formulated 49 requirements for occupational exoskeletons, and we validated and extended an existing framework. Our work encourages the improved design of future or existing exoskeletons.