Recent advancements in computational techniques significantly impact bioengineering, particularly in drug safety assessments and neuroscience trials using primate models. Macaques are extensively used due to their gen...
详细信息
In this work, we delve into the intricate synergy among non-prehensile actions like pushing, and prehensile actions such as grasping and throwing, within the domain of robotic manipulation. We introduce an innovative ...
详细信息
ISBN:
(数字)9798350384574
ISBN:
(纸本)9798350384581
In this work, we delve into the intricate synergy among non-prehensile actions like pushing, and prehensile actions such as grasping and throwing, within the domain of robotic manipulation. We introduce an innovative approach to learning these synergies by leveraging model-free deep reinforcement learning. The robot’s workflow involves detecting the pose of the target object and the basket at each time step, predicting the optimal push configuration to isolate the target object, determining the appropriate grasp configuration, and inferring the necessary parameters for an accurate throw into the basket. This empowers robots to skillfully reconfigure cluttered scenarios through pushing, creating space for collision-free grasping actions. Simultaneously, we integrate throwing behavior, showcasing how this action significantly extends the robot’s operational reach. Ensuring safety, we developed a simulation environment in Gazebo for robot training, applying the learned policy directly to our real robot. Notably, this work represents a pioneering effort to learn the synergy between pushing, grasping, and throwing actions. Extensive experimentation in both simulated and real-robot scenarios substantiates the effectiveness of our approach across diverse settings. Our approach achieves a success rate exceeding 80% in both simulated and real-world scenarios. A video showcasing our experiments is available online at: https://***/q1l4BJVDbRw
In recent years, there has been a growing interest in data-driven intelligent fault diagnosis, focusing on extracting fault features, which has emerged as a critical and challenging aspect of fault diagnosis. However,...
详细信息
This paper presents a spectral domain registration-based visual servoing scheme that works on 3D point clouds. Specifically, we propose a 3D model/point cloud alignment method, which works by finding a global transfor...
This paper presents a spectral domain registration-based visual servoing scheme that works on 3D point clouds. Specifically, we propose a 3D model/point cloud alignment method, which works by finding a global transformation between reference and target point clouds using spectral analysis. A 3D Fast Fourier Transform (FFT) in $\mathbb{R}^{3}$ is used for the translation estimation, and the real spherical harmonics in $\boldsymbol{SO}(3)$ are used for the rotations estimation. Such an approach allows us to derive a decoupled 6 degrees of freedom (DoF) controller, where we use gradient ascent optimisation to minimise translation and rotational costs. We then show how this methodology can be used to regulate a robot arm to perform a positioning task. In contrast to the existing state-of-the-art depth-based visual servoing methods that either require dense depth maps or dense point clouds, our method works well with partial point clouds and can effectively handle larger transformations between the reference and the target positions. Furthermore, the use of spectral data (instead of spatial data) for transformation estimation makes our method robust to sensor-induced noise and partial occlusions. We validate our approach by performing experiments using point clouds acquired by a robot-mounted depth camera. Obtained results demonstrate the effectiveness of our visual servoing approach.
Monocular visual odometry is an indispensable technique in visual signal tasks, including robotics and navigation. However, the scale ambiguity and poor generalizability of monocular systems remain significant challen...
详细信息
ISBN:
(数字)9798350359312
ISBN:
(纸本)9798350359329
Monocular visual odometry is an indispensable technique in visual signal tasks, including robotics and navigation. However, the scale ambiguity and poor generalizability of monocular systems remain significant challenges. To address this issue, we propose a novel framework to enhance visual odometry (VO) using monocular depth estimation algorithms and geometry-based methods. Our approach employs a depth point filtering method that utilizes depth information output from multiple depth estimation models with a minimum depth difference strategy. This approach effectively combines depth information from different models, resulting in more accurate scale factor recovery and trajectory prediction using PnP. Experiment results demonstrate that our method outperforms current learning-based VO methods in terms of generalization capability and accuracy. Additionally, we show that incorporating pre-trained models in the VO system to obtain pseudo-labeled depth data significantly enhances the performance of existing geometry-based VO methods.
Activities of daily living such as drinking and eating can be severely impaired for patients suffering from neurodegenerative diseases. One promising solution are assistive devices that apply corrective forces while s...
详细信息
ISBN:
(数字)9798350386523
ISBN:
(纸本)9798350386530
Activities of daily living such as drinking and eating can be severely impaired for patients suffering from neurodegenerative diseases. One promising solution are assistive devices that apply corrective forces while still allowing the intended movements. However, real-time estimation of the required forces requires a detailed understanding of the limb's impedance characteristics. Here, we test and validate the stiffness response of a computationally efficient neuro-musculoskeletal arm model and its response to various force perturbations. We demonstrate that the arm model predicts stiffness characteristics that closely match experimental data recorded from humans and presents real-time applicability, allowing for implementation in practical scenarios and. Additionally, we predict the stiffness response for novel force levels and arm configurations. In the future, these predictions could be used to estimate corrective forces for assistive devices in real-time.
Causal discovery from observational data is a crucial but challenging task, and learning directed acyclic graphs (DAGs) is its foundation. The causal discovery method in linear structural equation models (SEMs) is one...
详细信息
ISBN:
(数字)9798350372694
ISBN:
(纸本)9798350372700
Causal discovery from observational data is a crucial but challenging task, and learning directed acyclic graphs (DAGs) is its foundation. The causal discovery method in linear structural equation models (SEMs) is one of the hot spots. However, almost all the existing methods have certain limitations, and an exact solution cannot always be identified. In this paper, a new heuristic algorithm for discovering causality was proposed, which provides a reasonable solution for combining structured priors or possible expert knowledge with heuristic search. We first identify the partial v-structures through partial correlation analysis as the structural priors of the following heuristic search algorithm. Second, through partial correlation analysis, we can also limit the search space we want to search for. Finally, to adapt to structural priors, an efficient particle swarm optimization (PSO) algorithm with an enhanced local search and a dynamic double-swarm strategy is proposed to improve the search capability. The experimental results demonstrate the effectiveness of the proposed methods when compared to the earlier state-of-the-art methods on six standard networks.
Plasticine shape manipulation (PSM) is essential in food processing and ceramic production. However, it is challenging since high degrees of freedom of plasticine make it difficult for state representation and plannin...
详细信息
ISBN:
(数字)9798350373172
ISBN:
(纸本)9798350373189
Plasticine shape manipulation (PSM) is essential in food processing and ceramic production. However, it is challenging since high degrees of freedom of plasticine make it difficult for state representation and planning. Some previous approaches focus on optimizing actions using gradients provided by differentiable particle-based simulators, which still have a noticeable gap compared to real scenarios. In this paper, we use a long roller and present a specific inverse dynamics model (Plasti-Former) to perform PSM. The input to Plasti-Former consists of the provided sub-goals and the current point cloud state of the plasticine; the output is the robot's actions. Plasti-Former comprises a rolling direction prediction module and a rolling position prediction module. In the first module, we integrate optimal transport theory with a PointNet-based model trained with 30k virtual data and 6k actual data to predict the roller's rolling direction θ. In the second module, we predict the start rolling position (x, y) using an attention-based model. Note that the input of this attention-based model is a discretized spatial sequence uniquely designed by us, and the features of each subspace are constructed using the optimal transport vectors. We experimentally evaluate Plasti-Former on six representative sequential plasticine shape manipulation tasks and show that Plasti-Former outperforms state-of-the-art approaches in simu-lation. We also conducted the experiments in the real world, demonstrating a promising sim-to-real effect.
Teleoperation has the potential to enable robots to replace humans in high-risk scenarios and catastrophic events, performing manipulation tasks efficiently and securely under human guidance. However, achieving human-...
详细信息
As people's requirements for indoor positioning accuracy continue to increase, indoor wireless positioning technology has developed rapidly. Among the many wireless positioning technologies, ultra-wideband (UWB) w...
详细信息
ISBN:
(数字)9798350380323
ISBN:
(纸本)9798350380330
As people's requirements for indoor positioning accuracy continue to increase, indoor wireless positioning technology has developed rapidly. Among the many wireless positioning technologies, ultra-wideband (UWB) wireless positioning has quickly become a hot topic in today's indoor positioning research due to its strong anti-multipath interference ability, low power consumption, and high positioning accuracy. However, due to the complex and changeable indoor environment, UWB devices are interfered with by various factors in this environment, resulting in the inevitable NLOS (non-line-of-sight) error in the wireless positioning system, which hugely affects the positioning accuracy of the positioning system. For the problem of NLOS propagation in indoor pedestrian positioning due to received signal strength indication (RSSI), this paper proposes a locating algorithm that predicts RSSI using a motion model and corrects NLOS measurements. Firstly, the pedestrian's position and RSSI are predicted using a motion model, and NLOS errors are detected and corrected. Then, the pedestrian's activity range is estimated using step length prediction, and an improved trilateration algorithm is used for positioning. Finally, the Fang-unscented Kalman filter is applied to optimize the positioning results and obtain an optimized estimation of the position. Simulation experiments indicate that this method can validly improve locating accuracy, suppress large NLOS errors, and is an effective positioning method in NLOS environments.
暂无评论