Supernumerary robotic arms(SuperLimb)are a new type of wearable robot that works closely with humans as a third hand to augment human operation *** conveyance of wearers'intentions,allocation of roles,and humancen...
详细信息
Supernumerary robotic arms(SuperLimb)are a new type of wearable robot that works closely with humans as a third hand to augment human operation *** conveyance of wearers'intentions,allocation of roles,and humancentered interaction considerations are the key points in the process of human-SuperLimb *** paper proposes a human-centered intention-guided leader-follower controller that relies on the dynamic modeling of SuperLimb with application to load-carrying *** proposed leader-follower controller takes the human as the leader and the SuperLimb as the follower,achieving effective information communication,autonomous coordination,and good force compliance between SuperLimb,humans,and the environment under human safety ***,the human-SuperLimb dynamic system is modeled to achieve force interaction with the environment and ***,to achieve the precise intention extraction of humans,pose data from five visual odometry sensors are fused to capture the human state,the generalized position,the velocity of hands,and the surface electromyography signals from two myoelectric bracelets sensors are processed to recognize the natural hand gestures during load-carrying scenarios by a designed Swin transformer ***,based on the real-time distance detection between human and mechanical limbs,the security assurance and force-compliant interaction of the human-SuperLimb system are ***,the human hand muscle intention recognition,human-robot safety strategy verification,and comparative load-carrying experiments with and without the proposed method are conducted on the SuperLimb *** showed that the task parameters are well estimated to produce more reasonable planning trajectories,and SuperLimb could well understand the wearer's intentions to switch different SuperLimb *** proposed sensor-based human-robot communication framework motivates future studies of other collaboration scenes fo
Traditionally performed by skilled nozzle opera-tors, shotcrete remains a painful and risky task, where recurrent physical discomfort leads to severe and long-term health issues. When performed by novices or ill-train...
详细信息
ISBN:
(数字)9798350386523
ISBN:
(纸本)9798350386530
Traditionally performed by skilled nozzle opera-tors, shotcrete remains a painful and risky task, where recurrent physical discomfort leads to severe and long-term health issues. When performed by novices or ill-trained workers, it can lead to waste. The shotcrete process guidelines advise operators on how to best maneuver the nozzle tool during spraying to reduce waste. This technique involves moving the nozzle with rhythmic circular movements while remaining at an optimal distance from the shotcrete surface. Adhering to these standards is challenging and adds to the operator's discomfort. To alleviate these issues, robotic solutions can be developed to partially or completely substitute the operator's work. This paper presents a first step towards modeling the technique employed by experienced nozzle operators and transferring them to control a robotic arm. We recorded the motion generated at the nozzle during a set of shotcrete operations by two expert nozzle operators and one novice. We analyze the pattern of motion and confirm that we model the shotcrete task and showcase its use for autonomous control of a mock-up of a shotcrete robot using expandable foam. Furthermore, we implement an intuitive shared control framework to support operators during shotcrete. The experimental results from quasi-real-world evaluations of our proposed framework on a seven-degrees-of-freedom robotic manipulator demonstrate the efficacy of our proposed control approach.
In this article, we propose a dynamical system to avoid obstacles which are star shaped and simultaneously converge to a goal. The convergence is almost-global in a domain and the stationary points are identified expl...
详细信息
ISBN:
(数字)9783907144107
ISBN:
(纸本)9798331540920
In this article, we propose a dynamical system to avoid obstacles which are star shaped and simultaneously converge to a goal. The convergence is almost-global in a domain and the stationary points are identified explicitly. Our approach is based on the idea that an ideal vector field which avoids the obstacle traverses its boundary up to when a clear path to the goal is available. We show the existence of this clear path through a set connecting the boundary of the obstacle and the goal. The traversing vector field is determined for an arbitrary obstacle (de-scribed by a set of points) by separating it into cluster of stars. We propose an algorithm which is linear in number of points inside the obstacle. We verify the theoretical results presented with various hand drawn obstacle sets. Our methodology is also extended to obstacles which are not star-shaped, and, those which exist in high dimensions.
Established techniques that enable robots to learn from demonstrations are based on learning a stable dynamical system (DS). To increase the robots’ resilience to perturbations during tasks that involve static obstac...
详细信息
Machine learning models can solve complex tasks but often require significant computational resources during inference. This has led to the development of various post-training computation reduction methods that tackl...
详细信息
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...
详细信息
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.
Established techniques that enable robots to learn from demonstrations are based on learning a stable dynamical system (DS). To increase the robots’ resilience to perturbations during tasks that involve static obstac...
详细信息
ISBN:
(数字)9798350384574
ISBN:
(纸本)9798350384581
Established techniques that enable robots to learn from demonstrations are based on learning a stable dynamical system (DS). To increase the robots’ resilience to perturbations during tasks that involve static obstacle avoidance, we propose incorporating barrier certificates into an optimization problem to learn a stable and barrier-certified DS. Such optimization problem can be very complex or extremely conservative when the traditional linear parameter-varying formulation is used. Thus, different from previous approaches in the literature, we propose to use polynomial representations for DSs, which yields an optimization problem that can be tackled by sum-of-squares techniques. Finally, our approach can handle obstacle shapes that fall outside the scope of assumptions typically found in the literature concerning obstacle avoidance within the DS learning framework. Supplementary material can be found at the project webpage: https://***/abc-ds
While most approaches individually exploit unstructured data from the biomedical literature or structured data from biomedical knowl- edge graphs, their union can better exploit the advantages of such ap- proaches, ul...
详细信息
The inverse dynamics of a robotic manipulator is instrumental in precise robot control and manipulation. However, acquiring such a model is challenging, not only due to unmodelled non-linearities such as joint frictio...
详细信息
The inverse dynamics of a robotic manipulator is instrumental in precise robot control and manipulation. However, acquiring such a model is challenging, not only due to unmodelled non-linearities such as joint friction, but also from a machine learning perspective (e.g., input space dimension, amount of data needed). The accuracy of such models, regardless of the learning techniques, relies on proper excitation and exploration of the robot’s configuration space, in order to collect a rich dataset. This study aims to provide rich data in learning the inverse dynamics of a serial robotic manipulator using supervised machine learning techniques. We propose a method, called Max-Information Configuration Exploration (MICE), to incrementally explore and generate information-rich data via computing parameters of a trajectory set. We also introduce a new set of excitation trajectories that explores robot’s configuration through imposed stable limit cycles in robot joints’ phase space while satisfying feasibility constraints and physical bounds. We benchmark MICE against state-of-the-art in terms of data quality and learning accuracy. The proposed methodology for data collection, model learning, and evaluation, is validated with a KUKA IIWA14 robotic arm where the results prove significant improvement over traditional approaches.
We propose a structured prediction approach for robot imitation learning from demonstrations. Among various tools for robot imitation learning, supervised learning has been observed to have a prominent role. Structure...
详细信息
暂无评论