Urban water-surface robust perception serves as the foundation for intelligent monitoring of aquatic environments and the autonomous navigation and operation of unmanned vessels, especially in the context of waterway safety. It is worth noting that current multi-sensor fusion and multi-task learning models consume substantial power and heavily rely on high-power GPUs for inference. This contributes to increased carbon emissions, a concern that runs counter to the prevailing emphasis on environmental preservation and the pursuit of sustainable, low-carbon urban environments. In light of these concerns, this paper concentrates on low-power, lightweight, multi-task panoptic perception through the fusion of visual and 4D radar data, which is seen as a promising low-cost perception method. We propose a framework named Achelous++ that facilitates the development and comprehensive evaluation of multi-task water-surface panoptic perception models. Achelous++ can simultaneously execute five perception tasks with high speed and low power consumption, including object detection, object semantic segmentation, drivable-area segmentation, waterline segmentation, and radar point cloud semantic segmentation. Furthermore, to meet the demand for developers to customize models for real-time inference on low-performance devices, a novel multi-modal pruning strategy known as Heterogeneous-Aware SynFlow (HA-SynFlow) is proposed. Besides, Achelous++ also supports random pruning at initialization with different layer-wise sparsity, such as Uniform and Erdos-Renyi-Kernel (ERK). Overall, our Achelous++ framework achieves state-of-the-art performance on the WaterScenes benchmark, excelling in both accuracy and power efficiency compared to other single-task and multi-task models. We release and maintain the code at //github.com/GuanRunwei/Achelous.
相關內容
Recent rising interests in patient-specific thoracic surgical planning and simulation require efficient and robust creation of digital anatomical models from automatic medical image segmentation algorithms. Deep learning (DL) is now state-of-the-art in various radiological tasks, and U-shaped DL models have particularly excelled in medical image segmentation since the inception of the 2D UNet. To date, many variants of U-shaped models have been proposed by the integration of different attention mechanisms and network configurations. Leveraging the recent development of large multi-label databases, systematic benchmark studies for these models can provide valuable insights for clinical deployment and future model designs, but such studies are still rare. We conduct the first benchmark study for variants of 3D U-shaped models (3DUNet, STUNet, AttentionUNet, SwinUNETR, FocalSegNet, and a novel 3D SwinUnet with four variants) with a focus on CT-based anatomical segmentation for thoracic surgery. Our study systematically examines the impact of different attention mechanisms, number of resolution stages, and network configurations on segmentation accuracy and computational complexity. To allow cross-reference with other recent benchmarking studies, we also included a performance assessment of the BTCV abdominal structural segmentation. With the STUNet ranking at the top, our study demonstrated the value of CNN-based U-shaped models for the investigated tasks and the benefit of residual blocks in network configuration designs to boost segmentation performance.
Techniques of computer systems that have been successfully deployed for dense regular workloads fall short of achieving their goals of scalability and efficiency when applied to irregular and dynamic applications. This is primarily due to the discontent between the multiple layers of the system design from hardware architecture, execution model, programming model, to data-structure and application code. The paper approaches this issue by addressing all layers of the system design. It presents and argues key design principles needed for scalable and efficient dynamic graph processing, and from which it builds: 1) a fine-grain memory driven architecture that supports asynchronous active messages, 2) a programming and execution model that allows spawning tasks from within the data-parallelism, 3) and a data-structure that parallelizes vertex object across many compute cells and yet provides a single programming abstraction to the data object. Simulated experimental results show performance gain of geomean $2.38 \times$ against an state-of-the-art similar system for graph traversals and yet being able to natively support dynamic graph processing. It uses programming abstractions of actions, introduces new dynamic graph storage scheme, and message delivery mechanisms with continuations that contain post-completion actions. Continuations seamlessly adjusts, prior or running, execution to mutations in the input graph and enable dynamic graph processing.
Language-conditioned robotic manipulation represents a cutting-edge area of research, enabling seamless communication and cooperation between humans and robotic agents. This field focuses on teaching robotic systems to comprehend and execute instructions conveyed in natural language. To achieve this, the development of robust language understanding models capable of extracting actionable insights from textual input is essential. In this comprehensive survey, we systematically explore recent advancements in language-conditioned approaches within the context of robotic manipulation. We analyze these approaches based on their learning paradigms, which encompass reinforcement learning, imitation learning, and the integration of foundational models, such as large language models and vision-language models. Furthermore, we conduct an in-depth comparative analysis, considering aspects like semantic information extraction, environment & evaluation, auxiliary tasks, and task representation. Finally, we outline potential future research directions in the realm of language-conditioned learning for robotic manipulation, with the topic of generalization capabilities and safety issues. The GitHub repository of this paper can be found at //github.com/hk-zh/language-conditioned-robot-manipulation-models
For autonomous mobile robots, uncertainties in the environment and system model can lead to failure in the motion planning pipeline, resulting in potential collisions. In order to achieve a high level of robust autonomy, these robots should be able to proactively predict and recover from such failures. To this end, we propose a Gaussian Process (GP) based model for proactively detecting the risk of future motion planning failure. When this risk exceeds a certain threshold, a recovery behavior is triggered that leverages the same GP model to find a safe state from which the robot may continue towards the goal. The proposed approach is trained in simulation only and can generalize to real world environments on different robotic platforms. Simulations and physical experiments demonstrate that our framework is capable of both predicting planner failures and recovering the robot to states where planner success is likely, all while producing agile motion.
The emergence of Multimodal Large Language Models ((M)LLMs) has ushered in new avenues in artificial intelligence, particularly for autonomous driving by offering enhanced understanding and reasoning capabilities. This paper introduces LimSim++, an extended version of LimSim designed for the application of (M)LLMs in autonomous driving. Acknowledging the limitations of existing simulation platforms, LimSim++ addresses the need for a long-term closed-loop infrastructure supporting continuous learning and improved generalization in autonomous driving. The platform offers extended-duration, multi-scenario simulations, providing crucial information for (M)LLM-driven vehicles. Users can engage in prompt engineering, model evaluation, and framework enhancement, making LimSim++ a versatile tool for research and practice. This paper additionally introduces a baseline (M)LLM-driven framework, systematically validated through quantitative experiments across diverse scenarios. The open-source resources of LimSim++ are available at: //pjlab-adg.github.io/limsim_plus/.
Reconfigurable intelligent surfaces (RISs) have become a promising technology to meet the requirements of energy efficiency and scalability in future six-generation (6G) communications. However, a significant challenge in RISs-aided communications is the joint optimization of active and passive beamforming at base stations (BSs) and RISs respectively. Specifically, the main difficulty is attributed to the highly non-convex optimization space of beamforming matrices at both BSs and RISs, as well as the diversity and mobility of communication scenarios. To address this, we present a greenly gradient based meta learning beamforming (GMLB) approach. Unlike traditional deep learning based methods which take channel information directly as input, GMLB feeds the gradient of sum rate into neural networks. Coherently, we design a differential regulator to address the phase shift optimization of RISs. Moreover, we use the meta learning to iteratively optimize the beamforming matrices of BSs and RISs. These techniques make the proposed method to work well without requiring energy-consuming pre-training. Simulations show that GMLB could achieve higher sum rate than that of typical alternating optimization algorithms with the energy consumption by two orders of magnitude less.
Recent advancements in deep reinforcement learning (DRL) techniques have sparked its multifaceted applications in the automation sector. Managing complex decision-making problems with DRL encourages its use in the nuclear industry for tasks such as optimizing radiation exposure to the personnel during normal operating conditions and potential accidental scenarios. However, the lack of efficient reward function and effective exploration strategy thwarted its implementation in the development of radiation-aware autonomous unmanned aerial vehicle (UAV) for achieving maximum radiation protection. Here, in this article, we address these intriguing issues and introduce a deep Q-learning based architecture (RadDQN) that operates on a radiation-aware reward function to provide time-efficient minimum radiation-exposure pathway in a radiation zone. We propose a set of unique exploration strategies that fine-tune the extent of exploration and exploitation based on the state-wise variation in radiation exposure during training. Further, we benchmark the predicted path with grid-based deterministic method. We demonstrate that the formulated reward function in conjugation with adequate exploration strategy is effective in handling several scenarios with drastically different radiation field distributions. When compared to vanilla DQN, our model achieves a superior convergence rate and higher training stability.
Reconstructing the shape and spatially varying surface appearances of a physical-world object as well as its surrounding illumination based on 2D images (e.g., photographs) of the object has been a long-standing problem in computer vision and graphics. In this paper, we introduce an accurate and highly efficient object reconstruction pipeline combining neural based object reconstruction and physics-based inverse rendering (PBIR). Our pipeline firstly leverages a neural SDF based shape reconstruction to produce high-quality but potentially imperfect object shape. Then, we introduce a neural material and lighting distillation stage to achieve high-quality predictions for material and illumination. In the last stage, initialized by the neural predictions, we perform PBIR to refine the initial results and obtain the final high-quality reconstruction of object shape, material, and illumination. Experimental results demonstrate our pipeline significantly outperforms existing methods quality-wise and performance-wise.
Owing to effective and flexible data acquisition, unmanned aerial vehicle (UAV) has recently become a hotspot across the fields of computer vision (CV) and remote sensing (RS). Inspired by recent success of deep learning (DL), many advanced object detection and tracking approaches have been widely applied to various UAV-related tasks, such as environmental monitoring, precision agriculture, traffic management. This paper provides a comprehensive survey on the research progress and prospects of DL-based UAV object detection and tracking methods. More specifically, we first outline the challenges, statistics of existing methods, and provide solutions from the perspectives of DL-based models in three research topics: object detection from the image, object detection from the video, and object tracking from the video. Open datasets related to UAV-dominated object detection and tracking are exhausted, and four benchmark datasets are employed for performance evaluation using some state-of-the-art methods. Finally, prospects and considerations for the future work are discussed and summarized. It is expected that this survey can facilitate those researchers who come from remote sensing field with an overview of DL-based UAV object detection and tracking methods, along with some thoughts on their further developments.
The new era of technology has brought us to the point where it is convenient for people to share their opinions over an abundance of platforms. These platforms have a provision for the users to express themselves in multiple forms of representations, including text, images, videos, and audio. This, however, makes it difficult for users to obtain all the key information about a topic, making the task of automatic multi-modal summarization (MMS) essential. In this paper, we present a comprehensive survey of the existing research in the area of MMS.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191