Traditional farm autonomous feeding robot path planning systems have low target detection accuracy, poor planning efficiency, and insufficient adaptability to environmental changes when facing complex farm environment...
详细信息
ISBN:
(数字)9798350318609
ISBN:
(纸本)9798350318616
Traditional farm autonomous feeding robot path planning systems have low target detection accuracy, poor planning efficiency, and insufficient adaptability to environmental changes when facing complex farm environments. This article aims to solve the problems existing in traditional systems by introducing the YOLOv3 algorithm, in order to improve the performance and efficiency of robots in autonomous feeding tasks on farms. Firstly, this article utilizes the YOLOv3 algorithm to achieve target detection and localization in the system, and then uses sensor data fusion technology to perceive the environment and construct a map. It then integrates map information for path planning and optimization, achieving system adjustment and obstacle avoidance, and finally selects commonly used path planning algorithms for experimental comparison. The results show that the target detection accuracy of the task path planning system for autonomous feeding robots in three-dimensional farms based on the YOLOv3 algorithm is the highest, exceeding 99%. In addition, the operating efficiency of the system is also the highest, above 95%. In summary, the YOLOv3 algorithm is very suitable for the research of task path planning systems for autonomous feeding robots in three-dimensional farms.
Multidimensional parallel training has been widely applied to train large-scale deep learning models like GPT-3. The efficiency of parameter communication among training devices/processes is often the performance bott...
Multidimensional parallel training has been widely applied to train large-scale deep learning models like GPT-3. The efficiency of parameter communication among training devices/processes is often the performance bottleneck of large model training. Analysis of parameter communication mode and traffic has important reference significance for the research of interconnection network design and computing task scheduling to improve the training performance. In this paper, we analyze the parametric communication modes in typical 3D parallel training (data parallelism, pipeline parallelism, and tensor parallelism), and model the traffic in different communication modes. Finally, taking GPT-3 as an example, we present the communication in its 3D parallel training.
Underwater Optical Wireless sensor Networks (UOWSNs) are gaining an increasing demand in industrial and commercial applications as they can achieve high-speed communication. However, prior arts concentrate on promotin...
Underwater Optical Wireless sensor Networks (UOWSNs) are gaining an increasing demand in industrial and commercial applications as they can achieve high-speed communication. However, prior arts concentrate on promoting the performance of UOWSNs, while the reliability issue has not been fully addressed. In this paper, we propose a novel reliable data delivery scheme based on a cluster structure. First, we determine the orientation of each sensor for directional optical communication, which aims to establish reliable next-hop links among sensors. We formalize such an orientation problem into a submodular function maximization problem and propose a greedy method with an approximation ratio guarantee to solve it. Then, a cluster head designation scheme is developed to improve the data delivery success rate while minimizing the number of cluster heads. Finally, extensive simulations are conducted to demonstrate the effectiveness of the proposed scheme. The results reveal that compared with other algorithms, the proposed scheme can ensure a data delivery success rate of over 98.5 % while only keeping 45.3% fewer cluster heads. Furthermore, test-bed experiments are carried out to verify the applicability of the proposed scheme in practical applications.
The development of Federated Learning (FL) offers an efficient Machine Learning (ML) approach with privacy protection to solve the data island issue in distributed Internet of Things (IoT). However, existing FL framew...
The development of Federated Learning (FL) offers an efficient Machine Learning (ML) approach with privacy protection to solve the data island issue in distributed Internet of Things (IoT). However, existing FL frameworks still suffer from invisible attacks in IoT environments, such as free-rider attacks, backdoor attacks, and model theft. In this paper, we propose a Verifiable Property Federated Learning (VPFL) framework to overcome the above invisible attacks. We present a black-box watermarking task distribution mechanism to prevent free-rider attacks by verifying the property of local models. Our adversarial fine-tuning embedding technique can not only eliminate backdoors in global models, but also simultaneously embed white-box watermarks into model parameters to prevent model theft. Comprehensive experimental evaluations demonstrate that our framework outperforms state-of-the-art schemes in terms of security and feasibility against invisible attacks.
Vehicles are equipped with various sensors such as LiDAR, which enable them to perceive the surrounding environment and enhance driver safety through advanced driver assistance systems. However, these sensors are limi...
详细信息
ISBN:
(数字)9798350386059
ISBN:
(纸本)9798350386066
Vehicles are equipped with various sensors such as LiDAR, which enable them to perceive the surrounding environment and enhance driver safety through advanced driver assistance systems. However, these sensors are limited by line-of-sight, preventing them from seeing beyond occlusions. One solution is to leverage the edge server which can collect and share perception data with other vehicles. Most existing research focuses on improve the performance of uploading perception data to the server, and the problem of perception dissemination remains largely unexplored, despite the challenges posed by the large volume of perception data and the limited wireless bandwidth. In this paper, we propose an edge-assisted relevance-aware perception dissemination system that collects perception data from multiple vehicles and selectively disseminates only the necessary data to appropriate vehicles. The necessity of dissem-ination is determined by evaluating the relevance of perception data, which quantifies the probability of potential collisions between corresponding objects. We then formulate and solve the relevance-aware perception dissemination problem whose goal is to maximize the relevance of disseminated data under bandwidth constraints. Extensive evaluation results demonstrate that our system can significantly enhance traffic safety while reducing the overall bandwidth consumption.
Flying Ad Hoc Networks (FANETs) need reliable intra-communication to work effectively during the missions they participate in, especially in contested environments. Furthermore, choosing shortest paths for all e2e lin...
详细信息
ISBN:
(数字)9781665495127
ISBN:
(纸本)9781665495134
Flying Ad Hoc Networks (FANETs) need reliable intra-communication to work effectively during the missions they participate in, especially in contested environments. Furthermore, choosing shortest paths for all e2e links causes over-utilized hot spot nodes in the network. Thus, these hot spots have increased traffic that leads to more collisions around them. Consequently, these central nodes become easy targets for malicious attackers whose aim is to harm the network’s performance as efficiently as possible. To address this, we propose a routing framework (CentAir) that utilizes eigenvector centrality measures of the network nodes to fine-tune their transmission power. Then, it employs D* Lite path-finding algorithm to determine optimal e2e routes under varying network load. As a result, CentAir achieves up to 132% more channel capacity of e2e links compared to shortest path routing algorithms. Additionally, it shows 41% less throughput degradation when jammers are present.
The rapid growth of wearable sensor technologies holds substantial promise for the field of personalized and context-aware Human Activity Recognition. Given the inherently decentralized nature of data sources within t...
The rapid growth of wearable sensor technologies holds substantial promise for the field of personalized and context-aware Human Activity Recognition. Given the inherently decentralized nature of data sources within this domain, the utilization of multi-agent systems with their inherent decentralization capabilities presents an opportunity to facilitate the development of scalable, adaptable, and privacy-conscious methodologies. This paper introduces a collaborative distributed learning approach rooted in multi-agent principles, wherein individual users of sensor-equipped devices function as agents within a distributed network, collectively contributing to the comprehensive process of learning and classifying human ac-tivities. In this proposed methodology, not only is the privacy of activity monitoring data upheld for each individual, eliminating the need for an external server to oversee the learning process, but the system also exhibits the potential to surmount the limitations of conventional centralized models and adapt to the unique attributes of each user. The proposed approach has been empirically tested on two publicly accessible human activity recognition datasets, specifically PAMAP2 and HARTH, across varying settings. The provided empirical results conclusively highlight the efficacy of inter-individual collaborative learning when contrasted with centralized configurations, both in terms of local and gdobal generalization.
This paper presents a distributed structured controller design method in the case of finite number of subsystems with sensor failures. The constraint on the number of sensor failures is represented by a cardinality co...
详细信息
With the rapid development of intelligent devices in wireless communication, the fifth generation (5G) mobile networks have limited high data rates, low latency, high avail-ability demands. The sixth-generation (6G) m...
详细信息
ISBN:
(数字)9781665495127
ISBN:
(纸本)9781665495134
With the rapid development of intelligent devices in wireless communication, the fifth generation (5G) mobile networks have limited high data rates, low latency, high avail-ability demands. The sixth-generation (6G) mobile network can use Digital-twin (DT) techniques to meet these demands. DT is the virtual representation of physical aspects such as 6G edge nodes. DT optimize the 6G edge nodes parameters using artificial intelligence (AI) and especially machine learning (ML) algorithms. However, AI and ML bring along privacy and security concerns. Therefore, user data must be protected from unauthorized persons during the 6G edge network recovery and expansion phases. In this paper, we proposed a new reliable Digital Twin-based 6G edge network recovery framework using Blockchain technology. We applied the Transfer Learning (TL) technique to improve our proposed framework’s performance. We ensured data privacy and security using TL and Blockchain.
IoT and cyber-physical nodes with long operational lifetimes will be updated with some frequency after deployment. However, nodes deployed to environments like the arctic tundra are faced with several resource constra...
详细信息
ISBN:
(数字)9781665495127
ISBN:
(纸本)9781665495134
IoT and cyber-physical nodes with long operational lifetimes will be updated with some frequency after deployment. However, nodes deployed to environments like the arctic tundra are faced with several resource constraints, including a very small energy budget. To reduce energy consumption, the nodes are mostly sleeping. This makes updates more difficult because the update procedure can be interrupted by the nodes deciding to sleep or shut down. In addition, the updating system itself should be energy *** report on an approach and a prototype system for a single phase of the updating, concerned only with the problem of detecting that an update has arrived locally at a *** nodes can expect to receive updates at different update schedules. A series of performance measuring experiments were conducted on the update detection system to document how it behaves for different update *** conclude that to find a sweet spot for different update schedules with regard to the number of activations of the detection mechanism, and by implication also the energy consumption, and the time it takes before and update is detected, the update detection system must adapt to the different update schedules. While several factors impact this, the most significant effect comes from varying the time between activations of the detection mechanism.
暂无评论