Pub Date : 2024-09-18DOI: 10.1017/s0263574724001176
Seong Hyeon Hong, Benjamin Albia, Tristan Kyzer, Jackson Cornelius, Eric R. Mark, Asha J. Hall, Yi Wang
This paper presents an artificial neural network (ANN)-based nonlinear model predictive visual servoing method for mobile robots. The ANN model is developed for state predictions to mitigate the unknown dynamics and parameter uncertainty issues of the physics-based (PB) model. To enhance both the model generalization and accuracy for control, a two-stage ANN training process is proposed. In a pretraining stage, highly diversified data accommodating broad operating ranges is generated by a PB kinematics model and used to train an ANN model first. In the second stage, the test data collected from the actual system, which is limited in both the diversity and the volume, are employed to further finetune the ANN weights. Path-following experiments are conducted to compare the effects of various ANN models on nonlinear model predictive control and visual servoing performance. The results confirm that the pretraining stage is necessary for improving model generalization. Without pretraining (i.e., model trained only with the test data), the robot fails to follow the entire track. Weight finetuning with the captured data further improves the tracking accuracy by 0.07–0.15 cm on average.
本文介绍了一种基于人工神经网络（ANN）的移动机器人非线性模型预测视觉伺服方法。该人工神经网络模型用于状态预测，以缓解基于物理（PB）模型的未知动态和参数不确定性问题。为了提高模型的泛化能力和控制精度，提出了一个两阶段 ANN 训练过程。在预训练阶段，由 PB 运动学模型生成可适应广泛操作范围的高度多样化数据，并首先用于训练 ANN 模型。在第二阶段，利用从实际系统中收集到的测试数据来进一步微调 ANN 权重。我们进行了路径跟踪实验，以比较各种 ANN 模型对非线性模型预测控制和视觉伺服性能的影响。结果证实，预训练阶段对于提高模型泛化是必要的。如果不进行预训练（即仅使用测试数据训练模型），机器人将无法跟踪整个轨道。利用捕获的数据对权重进行微调，可进一步提高跟踪精度，平均提高 0.07-0.15 厘米。
{"title":"Artificial neural network-based model predictive visual servoing for mobile robots","authors":"Seong Hyeon Hong, Benjamin Albia, Tristan Kyzer, Jackson Cornelius, Eric R. Mark, Asha J. Hall, Yi Wang","doi":"10.1017/s0263574724001176","DOIUrl":"https://doi.org/10.1017/s0263574724001176","url":null,"abstract":"This paper presents an artificial neural network (ANN)-based nonlinear model predictive visual servoing method for mobile robots. The ANN model is developed for state predictions to mitigate the unknown dynamics and parameter uncertainty issues of the physics-based (PB) model. To enhance both the model generalization and accuracy for control, a two-stage ANN training process is proposed. In a pretraining stage, highly diversified data accommodating broad operating ranges is generated by a PB kinematics model and used to train an ANN model first. In the second stage, the test data collected from the actual system, which is limited in both the diversity and the volume, are employed to further finetune the ANN weights. Path-following experiments are conducted to compare the effects of various ANN models on nonlinear model predictive control and visual servoing performance. The results confirm that the pretraining stage is necessary for improving model generalization. Without pretraining (i.e., model trained only with the test data), the robot fails to follow the entire track. Weight finetuning with the captured data further improves the tracking accuracy by 0.07–0.15 cm on average.","PeriodicalId":49593,"journal":{"name":"Robotica","volume":"9 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142265239","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1017/s0263574724001371
Paraskevi Th. Zacharia, Elias K. Xidias
In various industrial robotic applications, the effective traversal of a manipulator amidst obstacles and its ability to reach specific task-points are imperative for the execution of predefined tasks. In certain scenarios, the sequence in which the manipulator reaches these task-points significantly impacts the overall cycle time required for task completion. Moreover, some tasks necessitate significant force exertion at the end-effector. Therefore, establishing an optimal sequence for the task-points reached by the end-effector’s tip is crucial for enhancing robot performance, ensuring collision-free motion and maintaining high-force application at the end-effector’s tip. To maximize the manipulator’s manipulability, which serves as a performance index for assessing its force capability, we aim to establish an optimal collision-free task sequence considering higher mechanical advantage. Three optimization criteria are considered: the cycle time, collision avoidance and the manipulability index. Optimization is accomplished using a genetic algorithm coupled with the Bump-Surface concept for collision avoidance. The effectiveness of this approach is confirmed through simulation experiments conducted in 2D and 3D environments with obstacles employing both redundant and non-redundant robots.
{"title":"Optimal robot task scheduling in cluttered environments considering mechanical advantage","authors":"Paraskevi Th. Zacharia, Elias K. Xidias","doi":"10.1017/s0263574724001371","DOIUrl":"https://doi.org/10.1017/s0263574724001371","url":null,"abstract":"In various industrial robotic applications, the effective traversal of a manipulator amidst obstacles and its ability to reach specific task-points are imperative for the execution of predefined tasks. In certain scenarios, the sequence in which the manipulator reaches these task-points significantly impacts the overall cycle time required for task completion. Moreover, some tasks necessitate significant force exertion at the end-effector. Therefore, establishing an optimal sequence for the task-points reached by the end-effector’s tip is crucial for enhancing robot performance, ensuring collision-free motion and maintaining high-force application at the end-effector’s tip. To maximize the manipulator’s manipulability, which serves as a performance index for assessing its force capability, we aim to establish an optimal collision-free task sequence considering higher mechanical advantage. Three optimization criteria are considered: the cycle time, collision avoidance and the manipulability index. Optimization is accomplished using a genetic algorithm coupled with the Bump-Surface concept for collision avoidance. The effectiveness of this approach is confirmed through simulation experiments conducted in 2D and 3D environments with obstacles employing both redundant and non-redundant robots.","PeriodicalId":49593,"journal":{"name":"Robotica","volume":"54 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142265241","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1017/s0263574724001115
Fenghua Wu, Wenbing Tang, Yuan Zhou, Shang-Wei Lin, Zuohua Ding, Yang Liu
Thanks to its real-time computation efficiency, deep reinforcement learning (DRL) has been widely applied in motion planning for mobile robots. In DRL-based methods, a DRL model computes an action for a robot based on the states of its surrounding obstacles, including other robots that may communicate with it. These methods always assume that the environment is attack-free and the obtained obstacles’ states are reliable. However, in the real world, a robot may suffer from obstacle localization attacks (OLAs), such as sensor attacks, communication attacks, and remote-control attacks, which cause the robot to retrieve inaccurate positions of the surrounding obstacles. In this paper, we propose a robust motion planning method ObsGAN-DRL, integrating a generative adversarial network (GAN) into DRL models to mitigate OLAs in the environment. First, ObsGAN-DRL learns a generator based on the GAN model to compute the approximation of obstacles’ accurate positions in benign and attack scenarios. Therefore, no detectors are required for ObsGAN-DRL. Second, by using the approximation positions of the surrounding obstacles, ObsGAN-DRL can leverage the state-of-the-art DRL methods to compute collision-free motion commands (e.g., velocity) efficiently. Comprehensive experiments show that ObsGAN-DRL can mitigate OLAs effectively and guarantee safety. We also demonstrate the generalization of ObsGAN-DRL.
{"title":"Robust motion planning for mobile robots under attacks against obstacle localization","authors":"Fenghua Wu, Wenbing Tang, Yuan Zhou, Shang-Wei Lin, Zuohua Ding, Yang Liu","doi":"10.1017/s0263574724001115","DOIUrl":"https://doi.org/10.1017/s0263574724001115","url":null,"abstract":"Thanks to its real-time computation efficiency, deep reinforcement learning (DRL) has been widely applied in motion planning for mobile robots. In DRL-based methods, a DRL model computes an action for a robot based on the states of its surrounding obstacles, including other robots that may communicate with it. These methods always assume that the environment is attack-free and the obtained obstacles’ states are reliable. However, in the real world, a robot may suffer from obstacle localization attacks (OLAs), such as sensor attacks, communication attacks, and remote-control attacks, which cause the robot to retrieve inaccurate positions of the surrounding obstacles. In this paper, we propose a robust motion planning method <jats:monospace>ObsGAN-DRL</jats:monospace>, integrating a generative adversarial network (GAN) into DRL models to mitigate OLAs in the environment. First, <jats:monospace>ObsGAN-DRL</jats:monospace> learns a generator based on the GAN model to compute the approximation of obstacles’ accurate positions in benign and attack scenarios. Therefore, no detectors are required for <jats:monospace>ObsGAN-DRL</jats:monospace>. Second, by using the approximation positions of the surrounding obstacles, <jats:monospace>ObsGAN-DRL</jats:monospace> can leverage the state-of-the-art DRL methods to compute collision-free motion commands (e.g., velocity) efficiently. Comprehensive experiments show that <jats:monospace>ObsGAN-DRL</jats:monospace> can mitigate OLAs effectively and guarantee safety. We also demonstrate the generalization of <jats:monospace>ObsGAN-DRL</jats:monospace>.","PeriodicalId":49593,"journal":{"name":"Robotica","volume":"11 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142265352","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1017/s0263574724000882
Jingzong Zhou, Priya Kulkarni, Xincheng Zhao, Sunil K. Agrawal
Traction of the head-neck is important in the treatment of patients suffering from neck pain due to degeneration of the intervertebral discs. Conventional neck traction is provided manually by experienced physical therapists who maintain a desired orientation of the head-neck relative to the trunk while applying the traction. It is postulated that innovative designs of neck exoskeletons can provide the same function both flexibly and accurately. This article presents a novel architecture of a parallel mechanism whose base sits on the human shoulders with 4 parallel chains, each chain having a revolute-revolute-universal-revolute (RRUR) structure, while the end-effector is connected rigidly to the human head. Each chain has five degrees-of-freedom (DOF) and applies a constraint on the motion of the end-effector. As a result, this parallel mechanism allows two DOFs to the end-effector. These are (i) forward flexion or lateral bending of the head and (ii) vertical translation. An important motivation for the current design with RRUR structure is to characterize the range of forward flexion/lateral bending of the head-neck with this structure and the vertical translation to the end-effector. A physical prototype was constructed and tested to evaluate the performance of this mechanism in hardware for the proposed application.
{"title":"Design of a traction neck brace with two degrees-of-freedom via a novel architecture of a spatial parallel mechanism","authors":"Jingzong Zhou, Priya Kulkarni, Xincheng Zhao, Sunil K. Agrawal","doi":"10.1017/s0263574724000882","DOIUrl":"https://doi.org/10.1017/s0263574724000882","url":null,"abstract":"Traction of the head-neck is important in the treatment of patients suffering from neck pain due to degeneration of the intervertebral discs. Conventional neck traction is provided manually by experienced physical therapists who maintain a desired orientation of the head-neck relative to the trunk while applying the traction. It is postulated that innovative designs of neck exoskeletons can provide the same function both flexibly and accurately. This article presents a novel architecture of a parallel mechanism whose base sits on the human shoulders with 4 parallel chains, each chain having a revolute-revolute-universal-revolute (<jats:italic>RRUR</jats:italic>) structure, while the end-effector is connected rigidly to the human head. Each chain has five degrees-of-freedom (DOF) and applies a constraint on the motion of the end-effector. As a result, this parallel mechanism allows two DOFs to the end-effector. These are (i) forward flexion or lateral bending of the head and (ii) vertical translation. An important motivation for the current design with <jats:italic>RRUR</jats:italic> structure is to characterize the range of forward flexion/lateral bending of the head-neck with this structure and the vertical translation to the end-effector. A physical prototype was constructed and tested to evaluate the performance of this mechanism in hardware for the proposed application.","PeriodicalId":49593,"journal":{"name":"Robotica","volume":"10 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142265240","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1017/s0263574724001358
Haonan Xi, Shaodong Li, Xi Liu
This work proposes a novel grasp detection method, the Efficient Grasp Aware Network (EGA-Net), for robotic visual grasp detection. Our method obtains semantic information for grasping through feature extraction. It efficiently obtains feature channel weights related to grasping tasks through the constructed ECA-ResNet module, which can smooth the network’s learning. Meanwhile, we use concatenation to obtain low-level features with rich spatial information. Our method inputs an RGB-D image and outputs the grasp poses and their quality score. The EGA-Net is trained and tested on the Cornell and Jacquard datasets, and we achieve 98.9% and 95.8% accuracy, respectively. The proposed method only takes 24 ms for real-time performance to process an RGB-D image. Moreover, our method achieved better results in the comparison experiment. In the real-world grasp experiments, we use a 6-degree of freedom (DOF) UR-5 robotic arm to demonstrate its robust grasping of unseen objects in various scenes. We also demonstrate that our model can successfully grasp different types of objects without any processing in advance. The experiment results validate our model’s exceptional robustness and generalization.
{"title":"A pixel-level grasp detection method based on Efficient Grasp Aware Network","authors":"Haonan Xi, Shaodong Li, Xi Liu","doi":"10.1017/s0263574724001358","DOIUrl":"https://doi.org/10.1017/s0263574724001358","url":null,"abstract":"This work proposes a novel grasp detection method, the Efficient Grasp Aware Network (EGA-Net), for robotic visual grasp detection. Our method obtains semantic information for grasping through feature extraction. It efficiently obtains feature channel weights related to grasping tasks through the constructed ECA-ResNet module, which can smooth the network’s learning. Meanwhile, we use concatenation to obtain low-level features with rich spatial information. Our method inputs an RGB-D image and outputs the grasp poses and their quality score. The EGA-Net is trained and tested on the Cornell and Jacquard datasets, and we achieve 98.9% and 95.8% accuracy, respectively. The proposed method only takes 24 ms for real-time performance to process an RGB-D image. Moreover, our method achieved better results in the comparison experiment. In the real-world grasp experiments, we use a 6-degree of freedom (DOF) UR-5 robotic arm to demonstrate its robust grasping of unseen objects in various scenes. We also demonstrate that our model can successfully grasp different types of objects without any processing in advance. The experiment results validate our model’s exceptional robustness and generalization.","PeriodicalId":49593,"journal":{"name":"Robotica","volume":"48 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142265242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Currently, there is an increasing demand for parallel mechanisms with fewer inputs and more outputs in aerospace, antennas, and recreational facilities. Therefore, this paper proposes a new parallel mechanism (PM) with a separation of constraints and drives, facilitating centralized drive management. Based on screw theory, the type synthesis of new parallel mechanisms with 3T (T representing translation), 3R (R representing revolute), 2T1R, and 2R1T is proposed. Three-degree-of-freedom (3-DOF) parallel mechanisms with fewer inputs and multiple outputs are proposed. Taking the 3RPS-PPPS parallel mechanism as an example, the closed-loop vector method is employed to solve for the lengths of actuated branches. A complete Jacobi matrix is constructed. The motion/force transmission performance of the 3RRS/PPPS parallel mechanism is better than the 3RRS parallel mechanism. Centralized management of drive branches greatly reduces the number of drives, making maintenance and repair easier. It has many application scenarios in antennas, recreational facilities, and other occasions.
{"title":"Type synthesis of a new parallel mechanism with complete separation of constraints and drives","authors":"Yu Wang, Can Qiu, Xiaoyu He, Jiabin Wu, Yundou Xu, Yongsheng Zhao","doi":"10.1017/s0263574724001346","DOIUrl":"https://doi.org/10.1017/s0263574724001346","url":null,"abstract":"Currently, there is an increasing demand for parallel mechanisms with fewer inputs and more outputs in aerospace, antennas, and recreational facilities. Therefore, this paper proposes a new parallel mechanism (PM) with a separation of constraints and drives, facilitating centralized drive management. Based on screw theory, the type synthesis of new parallel mechanisms with 3T (T representing translation), 3R (R representing revolute), 2T1R, and 2R1T is proposed. Three-degree-of-freedom (3-DOF) parallel mechanisms with fewer inputs and multiple outputs are proposed. Taking the 3RPS-PPPS parallel mechanism as an example, the closed-loop vector method is employed to solve for the lengths of actuated branches. A complete Jacobi matrix is constructed. The motion/force transmission performance of the 3RRS/PPPS parallel mechanism is better than the 3RRS parallel mechanism. Centralized management of drive branches greatly reduces the number of drives, making maintenance and repair easier. It has many application scenarios in antennas, recreational facilities, and other occasions.","PeriodicalId":49593,"journal":{"name":"Robotica","volume":"64 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142265353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1017/s0263574724000936
Sandro Augusto Costa Magalhães, Filipe Neves dos Santos, António Paulo Moreira, Jorge Manuel Miranda Dias
Performing tasks in agriculture, such as fruit monitoring or harvesting, requires perceiving the objects’ spatial position. RGB-D cameras are limited under open-field environments due to lightning interferences. So, in this study, we state to answer the research question: “How can we use and control monocular sensors to perceive objects’ position in the 3D task space?” Towards this aim, we approached histogram filters (Bayesian discrete filters) to estimate the position of tomatoes in the tomato plant through the algorithm MonoVisual3DFilter. Two kernel filters were studied: the square kernel and the Gaussian kernel. The implemented algorithm was essayed in simulation, with and without Gaussian noise and random noise, and in a testbed at laboratory conditions. The algorithm reported a mean absolute error lower than 10 mm in simulation and 20 mm in the testbed at laboratory conditions with an assessing distance of about 0.5 m. So, the results are viable for real environments and should be improved at closer distances.
{"title":"MonoVisual3DFilter: 3D tomatoes’ localisation with monocular cameras using histogram filters","authors":"Sandro Augusto Costa Magalhães, Filipe Neves dos Santos, António Paulo Moreira, Jorge Manuel Miranda Dias","doi":"10.1017/s0263574724000936","DOIUrl":"https://doi.org/10.1017/s0263574724000936","url":null,"abstract":"Performing tasks in agriculture, such as fruit monitoring or harvesting, requires perceiving the objects’ spatial position. RGB-D cameras are limited under open-field environments due to lightning interferences. So, in this study, we state to answer the research question: “How can we use and control monocular sensors to perceive objects’ position in the 3D task space?” Towards this aim, we approached histogram filters (Bayesian discrete filters) to estimate the position of tomatoes in the tomato plant through the algorithm MonoVisual3DFilter. Two kernel filters were studied: the square kernel and the Gaussian kernel. The implemented algorithm was essayed in simulation, with and without Gaussian noise and random noise, and in a testbed at laboratory conditions. The algorithm reported a mean absolute error lower than 10 mm in simulation and 20 mm in the testbed at laboratory conditions with an assessing distance of about 0.5 m. So, the results are viable for real environments and should be improved at closer distances.","PeriodicalId":49593,"journal":{"name":"Robotica","volume":"22 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142269373","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1017/s0263574724001280
Tao Liu, Jintao Zhao, Junhao Huang
Autonomous underwater vehicles (AUVs) have played a pivotal role in advancing ocean exploration and exploitation. However, traditional AUVs face limitations when executing missions at minimal or near-zero forward velocities due to the ineffectiveness of their control surfaces, considerably constraining their potential applications. To address this challenge, this paper introduces an innovative vectored thruster system based on a 3RRUR parallel manipulator tailored for micro-sized AUVs. The incorporation of a vectored thruster enhances the performance of micro-sized AUVs when operating at minimal and low forward speeds. A comprehensive exploration of the kinematics of the thrust-vectoring mechanism has been undertaken through theoretical analysis and experimental validation. The findings from theoretical analysis and experimental confirmation unequivocally affirm the feasibility of the devised thrust-vectoring mechanism. The precise control of the vector device is studied using Physics-informed Neural Network and Model Predictive Control (PINN-MPC). Through the adoption of this pioneering thrust-vectoring mechanism rooted in the 3RRUR parallel manipulator, AUVs can efficiently and effectively generate the requisite motion for thrust-vectoring propulsion, overcoming the limitations of traditional AUVs and expanding their potential applications across various domains.
{"title":"Kinematic analysis and advanced control of a vectored thruster based on 3RRUR parallel manipulator for micro-size AUVs","authors":"Tao Liu, Jintao Zhao, Junhao Huang","doi":"10.1017/s0263574724001280","DOIUrl":"https://doi.org/10.1017/s0263574724001280","url":null,"abstract":"Autonomous underwater vehicles (AUVs) have played a pivotal role in advancing ocean exploration and exploitation. However, traditional AUVs face limitations when executing missions at minimal or near-zero forward velocities due to the ineffectiveness of their control surfaces, considerably constraining their potential applications. To address this challenge, this paper introduces an innovative vectored thruster system based on a 3RRUR parallel manipulator tailored for micro-sized AUVs. The incorporation of a vectored thruster enhances the performance of micro-sized AUVs when operating at minimal and low forward speeds. A comprehensive exploration of the kinematics of the thrust-vectoring mechanism has been undertaken through theoretical analysis and experimental validation. The findings from theoretical analysis and experimental confirmation unequivocally affirm the feasibility of the devised thrust-vectoring mechanism. The precise control of the vector device is studied using Physics-informed Neural Network and Model Predictive Control (PINN-MPC). Through the adoption of this pioneering thrust-vectoring mechanism rooted in the 3RRUR parallel manipulator, AUVs can efficiently and effectively generate the requisite motion for thrust-vectoring propulsion, overcoming the limitations of traditional AUVs and expanding their potential applications across various domains.","PeriodicalId":49593,"journal":{"name":"Robotica","volume":"47 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142269371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1017/s0263574724000948
Tie Liu, Dianchun Bai, Hongyu Yi, Hiroshi Yokoi
As a kind of lower-limb motor assistance device, the intelligent walking aid robot plays an essential role in helping people with lower-limb diseases to carry out rehabilitation walking training. In order to enhance the safety performance of the lower-limb walking aid robot, this study proposes a deep vision-based abnormal lower-limb gait prediction model construction method for the problem of abnormal gait prediction of patients’ lower limbs. The point cloud depth vision technique is used to acquire lower-limb motion data, and a multi-posture angular prediction model is trained using long and short-term memory networks to build a model of the user’s lower-limb posture characteristics during normal walking as well as a real-time lower-limb motion prediction model. The experimental results indicate that the proposed lower-limb abnormal behavior prediction model is able to achieve a 97.4% prediction rate of abnormal lower-limb movements within 150 ms. Additionally, the model demonstrates strong generalization ability in practical applications. This paper proposes further ideas to enhance the safety performance of lower-limb rehabilitation robot use for patients with lower-limb disabilities.
{"title":"Prediction of abnormal gait behavior of lower limbs based on depth vision","authors":"Tie Liu, Dianchun Bai, Hongyu Yi, Hiroshi Yokoi","doi":"10.1017/s0263574724000948","DOIUrl":"https://doi.org/10.1017/s0263574724000948","url":null,"abstract":"As a kind of lower-limb motor assistance device, the intelligent walking aid robot plays an essential role in helping people with lower-limb diseases to carry out rehabilitation walking training. In order to enhance the safety performance of the lower-limb walking aid robot, this study proposes a deep vision-based abnormal lower-limb gait prediction model construction method for the problem of abnormal gait prediction of patients’ lower limbs. The point cloud depth vision technique is used to acquire lower-limb motion data, and a multi-posture angular prediction model is trained using long and short-term memory networks to build a model of the user’s lower-limb posture characteristics during normal walking as well as a real-time lower-limb motion prediction model. The experimental results indicate that the proposed lower-limb abnormal behavior prediction model is able to achieve a 97.4% prediction rate of abnormal lower-limb movements within 150 ms. Additionally, the model demonstrates strong generalization ability in practical applications. This paper proposes further ideas to enhance the safety performance of lower-limb rehabilitation robot use for patients with lower-limb disabilities.","PeriodicalId":49593,"journal":{"name":"Robotica","volume":"2 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142265349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-18DOI: 10.1017/s0263574724000985
Meiyi Yang, Junlin Xiong, Youfu Li
Visual odometry (VO) is a key technology for estimating camera motion from captured images. In this paper, we propose a novel RGB-D visual odometry by constructing and matching features at the superpixel level that represents better adaptability in different environments than state-of-the-art solutions. Superpixels are content-sensitive and perform well in information aggregation. They could thus characterize the complexity of the environment. Firstly, we designed the superpixel-based feature SegPatch and its corresponding 3D representation MapPatch. By using the neighboring information, SegPatch robustly represents its distinctiveness in various environments with different texture densities. Due to the inclusion of depth measurement, the MapPatch constructs the scene structurally. Then, the distance between SegPatches is defined to characterize the regional similarity. We use the graph search method in scale space for searching and matching. As a result, the accuracy and efficiency of matching process are improved. Additionally, we minimize the reprojection error between the matched SegPatches and estimate camera poses through all these correspondences. Our proposed VO is evaluated on the TUM dataset both quantitatively and qualitatively, showing good balance to adapt to the environment under different realistic conditions.
视觉轨迹测量（VO）是根据捕捉到的图像估计摄像机运动的一项关键技术。在本文中，我们提出了一种新颖的 RGB-D 视觉里程测量法，它通过在超像素级别构建和匹配特征，与最先进的解决方案相比，在不同环境下具有更好的适应性。超像素对内容敏感，在信息聚合方面表现出色。因此，它们可以描述环境的复杂性。首先，我们设计了基于超像素的特征 SegPatch 及其相应的三维表示 MapPatch。通过使用邻近信息，SegPatch 可以在不同纹理密度的环境中稳健地表示其独特性。由于包含了深度测量，MapPatch 从结构上构建了场景。然后，定义 SegPatch 之间的距离来描述区域相似性。我们在比例空间中使用图搜索法进行搜索和匹配。因此，匹配过程的准确性和效率都得到了提高。此外，我们将匹配的 SegPatches 之间的重投影误差降至最低，并通过所有这些对应关系来估计摄像机的姿势。我们提出的 VO 在 TUM 数据集上进行了定量和定性评估，显示出在不同现实条件下适应环境的良好平衡性。
{"title":"RGB-D visual odometry by constructing and matching features at superpixel level","authors":"Meiyi Yang, Junlin Xiong, Youfu Li","doi":"10.1017/s0263574724000985","DOIUrl":"https://doi.org/10.1017/s0263574724000985","url":null,"abstract":"Visual odometry (VO) is a key technology for estimating camera motion from captured images. In this paper, we propose a novel RGB-D visual odometry by constructing and matching features at the superpixel level that represents better adaptability in different environments than state-of-the-art solutions. Superpixels are content-sensitive and perform well in information aggregation. They could thus characterize the complexity of the environment. Firstly, we designed the superpixel-based feature SegPatch and its corresponding 3D representation MapPatch. By using the neighboring information, SegPatch robustly represents its distinctiveness in various environments with different texture densities. Due to the inclusion of depth measurement, the MapPatch constructs the scene structurally. Then, the distance between SegPatches is defined to characterize the regional similarity. We use the graph search method in scale space for searching and matching. As a result, the accuracy and efficiency of matching process are improved. Additionally, we minimize the reprojection error between the matched SegPatches and estimate camera poses through all these correspondences. Our proposed VO is evaluated on the TUM dataset both quantitatively and qualitatively, showing good balance to adapt to the environment under different realistic conditions.","PeriodicalId":49593,"journal":{"name":"Robotica","volume":"36 1","pages":""},"PeriodicalIF":2.7,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142265351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：