检索结果-内蒙古大学图书馆

Pedestrian Trajectory Prediction Using Dynamics-based Deep Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Honghui Zhi, Weiming Batista, Gustavo Chandra, Rohitash Transitional Artificial Intelligence Research Group School of Mathematics and Statistics UNSW Sydney Sydney Australia Robotics Institute Carnegie Mellon University Pittsburgh United States School of Computer Science and Engineering UNSW Sydney Sydney Australia

Pedestrian trajectory prediction plays an important role in autonomous driving systems and robotics. Recent work utilizing prominent deep learning models for pedestrian motion prediction makes limited a priori assumptions about human movements, resulting in a lack of explainability and explicit constraints enforced on predicted trajectories. We present a dynamics-based deep learning framework with a novel asymptotically stable dynamical system integrated into a Transformer-based model. We use an asymptotically stable dynamical system to model human goal-targeted motion by enforcing the human walking trajectory, which converges to a predicted goal position, and to provide the Transformer model with prior knowledge and explainability. Our framework features the Transformer model that works with a goal estimator and dynamical system to learn features from pedestrian motion history. The results show that our framework outperforms prominent models using five benchmark human motion datasets. © 2023, CC BY.

关键词： Dynamical systems

Vision-Based Traffic Accident Detection and Anticipation: A Survey

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Fang, Jianwu Qiao, Jiahuan Xue, Jianru Li, Zhengguo College of Transportation Engineering Chang’an University Xi’an China Applied Mathematics and Artificial Intelligence Institute of Information Technologies and Computer Science National Research University Moscow Power Engineering Institute Russia Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an China Singapore

Traffic accident detection and anticipation is an obstinate road safety problem and painstaking efforts have been devoted. With the rapid growth of video data, Vision-based Traffic Accident Detection and Anticipation (named Vision-TAD and Vision-TAA) become the last one-mile problem for safe driving and surveillance safety. However, the long-tailed, unbalanced, highly dynamic, complex, and uncertain properties of traffic accidents form the Out-of-Distribution (OOD) feature for Vision-TAD and Vision-TAA. Current AI development may focus on these OOD but important problems. What has been done for Vision-TAD and Vision-TAA? What direction we should focus on in the future for this problem? A comprehensive survey is important. We present the first survey on Vision-TAD in the deep learning era and the first-ever survey for Vision-TAA. The pros and cons of each research prototype are discussed in detail during the investigation. In addition, we also provide a critical review of 31 publicly available benchmarks and related evaluation metrics. Through this survey, we want to spawn new insights and open possible trends for Vision-TAD and Vision-TAA tasks. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep learning

A Portable Multiscopic Camera for Novel View and Time Synthesis in Dynamic Scenes

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhang, Tianjia Lau, Yuen-Fui Chen, Qifeng The Hong Kong University of Science and Technology Hong Kong The Individualized Interdisciplinary Program The Robotics Institute Hong Kong The Department of Mathematics Hong Kong The Department of Computer Science and Engineering The Department of Electronic and Computer Engineering Hong Kong

We present a portable multiscopic camera system with a dedicated model for novel view and time synthesis in dynamic scenes. Our goal is to render high-quality images for a dynamic scene from any viewpoint at any time using our portable multiscopic camera. To achieve such novel view and time synthesis, we develop a physical multiscopic camera equipped with five cameras to train a neural radiance field (NeRF) in both time and spatial domains for dynamic scenes. Our model maps a 6D coordinate (3D spatial position, 1D temporal coordinate, and 2D viewing direction) to view-dependent and time-varying emitted radiance and volume density. Volume rendering is applied to render a photo-realistic image at a specified camera pose and time. To improve the robustness of our physical camera, we propose a camera parameter optimization module and a temporal frame interpolation module to promote information propagation across time. We conduct experiments on both real-world and synthetic datasets to evaluate our system, and the results show that our approach outperforms alternative solutions qualitatively and quantitatively. Our code and dataset are available at https://***/. Copyright © 2022, The Authors. All rights reserved.

关键词： Cameras

Embodied Runtime Monitoring of Learning-enabled Robot Perception Components

学校读者我要写书评

暂无评论

Embodied Runtime Monitoring of Learning-enabled Robot Percep...

IEEE International Conference on Automation science and Engineering (CASE)

作者： Deebul Nair Sathwik Panchangam Miguel A. Olivares-Mendez Nico Hochgeschwender Institute for Artificial Intelligence and Autonomous Systems Bonn-Rhein-Sieg University of Applied Sciences Sankt Augustin Germany Space Robotics Research Group (SpaceR) Interdisciplinary Centre for Security Reliability and Trust (SnT) University of Luxembourg Luxembourg Department of Mathematics and Computer Science University of Bremen Germany

ISBN: (数字)9798350358513

ISBN: (纸本)9798350358520

Monitoring during runtime, the estimated uncertainty from learning-enabled robot perception components, is a vital tool for assuring and evaluating model correctness. By utilizing different uncertainty estimation methods, monitors exploit the perceived sensor data and predictions generated by the perception model to make informed decisions, such as in safety-critical situations. In robotics, informed decisions should be aligned with embodied information regarding space and time. We argue and show that runtime monitors of learning-enabled components can be made more robust by including the embodiment information of the robot. However, it remains challenging to build uncertainty-based runtime monitors with embodiment information using only real-world datasets, which are tedious to collect. To this end, we propose a solution that leverages simulation-based datasets to characterize expected uncertainty estimates with robot embodiment. By constructing a nominal runtime monitor model based on simulated datasets, we establish a reference for the expected uncertainty along with the embodiment information that can be subsequently used to monitor real-world deployed perception components. We demonstrate the zero-shot sim-to-real transfer of our approach on a Kinova Gen3 manipulator picking task, for which the runtime monitor reliability model was trained using simulation-based data containing embodiment information of distance from the object being selected.

关键词： Runtime Uncertainty Computational modeling Estimation Predictive models Robot sensing systems Manipulators Reliability engineering Data models Monitoring

F2A2: flexible fully-decentralized approximate actor-critic for cooperative multi-agent reinforcement learning

学校读者我要写书评

暂无评论

The Journal of Machine Learning Research

The Journal of Machine Learning Research 2023年第1期24卷 8447-8521页

作者： Wenhao Li Bo Jin Xiangfeng Wang Junchi Yan Hongyuan Zha School of Data Science The Chinese University of Hong Kong Shenzhen Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China School of Software Engineering Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai China School of Computer Science and Technology Key Laboratory of Mathematics and Engineering Applications Ministry of Education East China Normal University Shanghai China Department of Computer Science and Engineering Key Laboratory of Artificial Intelligence Ministry of Education Shanghai Jiao Tong University Shanghai China

Traditional centralized multi-agent reinforcement learning (MARL) algorithms are sometimes unpractical in complicated applications due to non-interactivity between agents, the curse of dimensionality, and computation complexity. Hence, several decentralized MARL algorithms are motivated. However, existing decentralized methods only handle the fully cooperative setting where massive information needs to be transmitted in training. The block coordinate gradient descent scheme they used for successive independent actor and critic steps can simplify the calculation, but it causes serious bias. This paper proposes a exible fully decentralized actor-critic MARL framework, which can combine most of the actor-critic methods and handle large-scale general cooperative multi-agent settings. A primal-dual hybrid gradient descent type algorithm framework is designed to learn individual agents separately for decentralization. From the perspective of each agent, policy improvement and value evaluation are jointly optimized, which can stabilize multi-agent policy learning. Furthermore, the proposed framework can achieve scalability and stability for the large-scale environment. This framework also reduces information transmission by the parameter sharing mechanism and novel modeling-other-agents methods based on theory-of-mind and online supervised learning. Sufficient experiments in cooperative Multi-agent Particle Environment and StarCraft II show that the proposed decentralized MARL instantiation algorithms perform competitively against conventional centralized and decentralized methods.

关键词： cooperative MARL decentralized actor-critic primal-dual method

LMS-Net: A Learned Mumford-Shah Network For Few-Shot Medical Image Segmentation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhang, Shengdong Jia, Fan Li, Xiang Zhang, Hao Shi, Jun Ma, Liyan Ying, Shihui Department of Mathematics School of Science Shanghai University Shanghai200444 China Institute University of Utah Salt Lake CityUT84102 United States School of Computer Science and Technology East China Normal University Shanghai China School of Communication and Information Engineering Shanghai University Shanghai200444 China School of Computer Engineering and Science Shanghai University Shanghai200444 China School of Mechatronic Engineering and Automation Shanghai Key Laboratory of Intelligent Manufacturing and Robotics Shanghai University Shanghai200444 China Shanghai Institute of Applied Mathematics and Mechanics The School of Mechanics and Engineering Science Shanghai University Shanghai200072 China

Few-shot semantic segmentation (FSS) methods have shown great promise in handling data-scarce scenarios, particularly in medical image segmentation tasks. However, most existing FSS architectures lack sufficient interpretability and fail to fully incorporate the underlying physical structures of semantic regions. To address these issues, in this paper, we propose a novel deep unfolding network, called the Learned Mumford-Shah Network (LMSNet), for the FSS task. Specifically, motivated by the effectiveness of pixel-to-prototype comparison in prototypical FSS methods and the capability of deep priors to model complex spatial structures, we leverage our learned Mumford-Shah model (LMS model) as a mathematical foundation to integrate these insights into a unified framework. By reformulating the LMS model into prototype update and mask update tasks, we propose an alternating optimization algorithm to solve it efficiently. Further, the iterative steps of this algorithm are unfolded into corresponding network modules, resulting in LMS-Net with clear interpretability. Comprehensive experiments on three publicly available medical segmentation datasets verify the effectiveness of our method, demonstrating superior accuracy and robustness in handling complex structures and adapting to challenging segmentation scenarios. These results highlight the potential of LMS-Net to advance FSS in medical imaging applications. Our code will be available at: https:// ***/SDZhang01/LMSNet Copyright © 2025, The Authors. All rights reserved.

关键词： Semantic Segmentation

Offline Content-based Recommendation System for Wikimedia Commons Contents

学校读者我要写书评

暂无评论

Procedia computer science 2025年 257卷 485-494页

作者： Ahidjo Maitaigahasse Jean Louis Kedieng Ebongue Fendji Marcellin Atemkeng Department Mathematics and Computer Science Faculty of Science University of Ngaoundere 454 Ngaoundere Cameroon Centre for Research Experimentation and Production School of Chemical Engineering and Mineral Industries University of Ngaoundere 454 Ngaoundere Cameroon Stellenbosch Institute for Advanced Study (STIAS) Wallenberg Research Centre at Stellenbosch University Stellenbosch South Africa Department of Robotics and Industrial Informatics School of Chemical Engineering and Mineral Industries University of Ngaoundere 454 Ngaoundere Cameroon Department of Mathematics Rhodes University P.O. Box 6140 Grahamstown South Africa

Recommendation systems are essential for enhancing digital experiences, but their reliance on internet connectivity limits accessibility in regions with limited or no access. This paper presents an offline content-based recommendation system designed to operate without internet dependency by leveraging precomputed Term Frequency-Inverse Document Frequency (TF-IDF) vectors and cosine similarity. To optimize offline performance, we store a locally computed TF-IDF matrix, allowing efficient retrieval of relevant media items through matrix multiplication instead of real-time computation. The system is evaluated using twenty simulated personas representing diverse user interests, demonstrating its ability to generate personalized relevant recommendations. By eliminating the need for online data access, our system makes educational content from Wikimedia Commons accessible in remote areas, schools, and offline learning environments. These findings highlight the potential of offline recommendation systems in bridging the digital divide and providing equitable access to personalized learning resources.

关键词： Content-based recommendation system Wikimedia commons TF-IDF Cosine Similarity

Shape Sensing of a Soft Robot Using Fiber Bragg Gratings for Single-port Transvesical Radical Prostatectomy

学校读者我要写书评

暂无评论

Shape Sensing of a Soft Robot Using Fiber Bragg Gratings for...

IEEE International Conference on robotics and Biomimetics

作者： Zefeng Liu Ru Li Yongfeng Cao Fan Feng Le Xie School of Materials Science and Engineering and the Institute of Medical Robotics Shanghai Jiao Tong University Shanghai China School of Materials Sci-ence and Engineering and the Institute of Medical Robotics Shanghai Jiao Tong University Shanghai China School of Mathematics and Computer Science Quanzhou Normal University Quanzhou China Institute of Medical Robotics Institute of Forming Technology & Equipment Shanghai Jiao Tong University Shanghai China

ISBN: (纸本)9781665481106

Robot-assisted laparoscopic radical prostatectomy is a common treatment for prostate cancer. Traditional rigid instruments, on the other hand, have operating blind areas and lack real-time position and shape feedback, which may result in additional trauma during the surgery. To this end, this work describes the development of a soft robot for single-port transvesical radical prostatectomy (STRP) using a multicore fiber with Fiber Bragg Gratings (FBGs). The shape sensing method is proposed in the meanwhile to rebuild the soft robot's shape from the wavelength data of FBGs. The experiments are carried out to confirm that the suggested strategy is workable.

关键词： Shape Multicore processing Surgery Soft robotics Reconstruction algorithms Robot sensing systems Fiber gratings

Lightweight Small Target Detection Algorithm with Multi-Feature Fusion

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Yang, Rujin Zhang, Jingwei Shang, Xinna Li, Wenfa Zhu, Deping Beijing Key Laboratory of Information Service Engineering Beijing Union University Beijing100101 China College of Robotics Beijing Union University Beijing100101 China Institute of Artificial Intelligence University of Science and Technology Beijing Beijing100083 China Vorovich Institute of Mathematics Mechanics and Computer Science Southern Federal University 344090 Russia

Unmanned aerial vehicles (UAVs) are a highly sought-after technology that has a wide range of applications in both military and civilian fields. Target detection is a crucial aspect of UAV applications, but there are still challenges associated with complex detection models and difficulty in detecting small targets. To address these issues, this paper proposes the lightweight L-YOLO algorithm for target detection tasks from a UAV perspective. The L-YOLO algorithm is an improvement on YOLOv5 that enhances the detection performance of the model for small targets while reducing the number of parameters and computational effort. The GhostNet module replaces the relevant convolution in the YOLOv5 model to create a lightweight model. The EIoU Loss is used as the loss function of the algorithm to accelerate convergence and improve regression accuracy. In addition, feature-level extensions based on YOLOv5 are implemented, and a new detection head is proposed to enhance the detection accuracy of the model for small targets. The K-means++ clustering algorithm is used to redesign the size of the anchor box for the small target dataset, resolving the issue of unsuitable anchor boxes for small targets. The experiments were conducted on the VisDrone-2022 dataset, and the L-YOLO algorithm demonstrated a reduction in computational effort by 42.42% and a decrease in the number of parameters by 48.6% compared to the original algorithm. Furthermore, recall improved by 2.1% and mAP@0.5 improved by 1.4%. These results demonstrate that the L-YOLO algorithm not only has better detection performance for small targets but also a lighter model, indicating promising prospects for target detection from a UAV perspective. © 2023, The Authors. All rights reserved.

关键词： Antennas