检索结果-内蒙古大学图书馆

2023 Australasian Conference on robotics and automation, ACRA 2023

作者： Liu, Violet Chen, Jason Qureshi, Ans Nejati, Mahla Centre for Automation and Robotic Engineering Science New Zealand The University of Auckland New Zealand

Amidst growing food production demands, early plant disease detection is essential to safeguard crops;this study proposes a visual machine learning approach for plant disease detection, harnessing RGB and NIR data collected in real-world conditions through a JAI FS-1600D10GE camera to build an RGBN dataset. A two-stage early plant disease detection model with YOLOv8 and a sequential CNN was used to train on a dataset with partial labels, which showed a 3.6% increase in mAP compared to a single-stage end-to-end segmentation model. The sequential CNN model achieved 90.62% validation accuracy utilising RGBN data. An average of 6.25% validation accuracy increase is found using RGBN in classification compared to RGB using ResNet15 and the sequential CNN models. Further research and dataset improvements are needed to meet food production demands. © 2023 Australasian robotics and automation Association. All rights reserved.

关键词： robotics

来源：评论

学校读者我要写书评

暂无评论

Textile Anomaly Detection: Evaluation of the State-of-the-Art for Automated Quality Inspection of Carpet

Textile Anomaly Detection: Evaluation of the State-of-the-Ar...

引用

2023 Australasian Conference on robotics and automation, ACRA 2023

作者： Forsberg, Briony Williams, Henry MacDonald, Bruce Chen, Tracy Hulse, Kirstine Centre for Automation and Robotic Engineering Science The University of Auckland New Zealand Bremworth Ltd Auckland New Zealand

In this study, state-of-the-art unsupervised detection models were evaluated for the purpose of automated anomaly inspection of wool carpets. A custom dataset of four unique types of carpet textures was created to thoroughly test the models and their robustness in detecting subtle anomalies in complex textures. Due to the requirements of an inline inspection system in a manufacturing use case, the metrics of importance in this study were accuracy in detecting anomalous areas, the number of false detections, and the inference times of each model for real-time performance. Of the evaluated models, the student-teacher network based methods were found on average to yield the highest detection accuracy and lowest false detection rates. When trained on a multi-class dataset the models were found to yield comparable if not better results than single-class training. Finally, in terms of detection speed, with exception to the generative model, all other evaluated models were found to have comparable inference times on a GPU, with an average of 0.16s per image. On a CPU, most of these models typically produced results between 1.5 to 2 times the respective GPU inference times. © 2023 Australasian robotics and automation Association. All rights reserved.

关键词： Textures

来源：评论

学校读者我要写书评

暂无评论

The Effect of Dimensional Parameters on the Performance of DDMRs 19

The Effect of Dimensional Parameters on the Performance of D...

引用

19th IEEE International Conference on automation science and engineering, CASE 2023

作者： Tattersall, Andrew McGuinness, Benjamin Lim, Hin Au, Chi Kit Williams, Henry Duke, Mike The School of Engineering University of Waikato Hamilton New Zealand The Centre for Automation and Robotic Engineering Science The University of Auckland New Zealand

ISBN: (纸本)9798350320695

Differential drive mobile robots (DDMRs) are often used in automation research because they are simple mechanically, and low cost as they only require two motors to drive and steer. Work to date has focused on control, with little done to optimise these vehicles in terms of dimensional parameters such as centre of mass position, track width, wheel base and caster dimensions. This paper presents a dynamic model of a DDMR and the dimensional sensitivity of peak drive/braking force, energy consumption, and chassis forces is assessed. The model is applied to a range of trajectories typical in an agricultural setting. Results show that these are all sensitive to vehicle dimensions to different degrees. Energy consumption showed the least sensitivity, while drive forces were found to be significantly influenced, for example, increasing the caster trail from 0.13 to 0.33 m decreased peak drive force by 56%. Optimising dimensional parameters in DDMRs can contribute to more economically viable automated solutions through increased efficency and reduced expense. © 2023 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Redefining Selective Blueberry Harvesting - Localised Shaking with Minimal Damage

Redefining Selective Blueberry Harvesting - Localised Shakin...

引用

2023 Australasian Conference on robotics and automation, ACRA 2023

作者： Sim, Alicia McGuinness, Ben Lim, Shen Hin Williams, Henry Duke, Mike the School of Engineering University of Waikato Hamilton New Zealand the Centre for Automation and Robotic Engineering Science The University of Auckland New Zealand

This paper presents a harvesting system that utilises localised shaking combined with a multi-layer soft catchment device for harvesting ripe blueberries. This paper also introduces a harvesting evaluation process to quantify the efficiency of the developed system. During preliminary testing within a commercial blueberry farm, the device showed a harvesting efficiency of 64% for Eureka and Masena varieties. This is lower in efficiency than existing manual harvesting methods into buckets, which were between 82.7 and 89.3 %. However, the device has not been optimised, and testing was limited. The devices show promise;further developments and trials are required to optimise the shaking frequency to maximise harvest efficiency and determine the harvesting rates. © 2023 Australasian robotics and automation Association. All rights reserved.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Seeing the Fruit for the Leaves: Towards Automated Apple Fruitlet Thinning

Seeing the Fruit for the Leaves: Towards Automated Apple Fru...

引用

2022 Australasian Conference on robotics and automation, ACRA 2022

作者： Qureshi, Ans Loh, Neville Kwon, Young Min Smith, David Gee, Trevor Bachelor, Oliver McCulloch, Josh Nejati, Mahla Lim, JongYoon Green, Richard Ahn, Ho Seok MacDonald, Bruce Williams, Henry Centre for Automation and Robotic Engineering Science University of Auckland New Zealand Department of Computer Science and Software Engineering University of Canterbury New Zealand

ISBN: (纸本)9781713866480

Following a global trend, the lack of reliable access to skilled labour is causing critical issues for the effective management of apple orchards. One of the primary challenges is maintaining skilled human operators capable of making precise fruitlet thinning decisions. Thinning requires accurately measuring the true crop load for individual apple trees to provide optimal thinning decisions on an individual basis. A challenging task due to the dense foliage obscuring the fruitlets within the tree structure. This paper presents the initial design, implementation, and evaluation details of the vision system for an automatic apple fruitlet thinning robot to meet this need. The platform consists of a UR5 robotic arm and stereo cameras which enable it to look around the leaves to map the precise number and size of the fruitlets on the apple branches. We show that this platform can measure the fruitlet load on the apple tree to with 84% accuracy in a real-world commercial apple orchard while being 87% precise. © 2022 Australasian robotics and automation Association. All rights reserved.

关键词： Fruits

来源：评论

学校读者我要写书评

暂无评论

On the Efficient Learning and Execution of Real-World, Dexterous In-Hand Manipulation Tasks with Phased Behaviour-Guiding Reward

On the Efficient Learning and Execution of Real-World, Dexte...

引用

2024 Australasian Conference on robotics and automation, ACRA 2024

作者： Xing, Yuning van Rijnsoever, Koen Valencia, David Gee, Trevor MacDonald, Bruce A. Liarokapis, Minas Williams, Henry Centre for Automation and Robotic Engineering Science The University of Auckland New Zealand New Dexterity research group The University of Auckland New Zealand

Reinforcement Learning (RL) training typically occurs in simulations using standardized, simple reward functions, often requiring millions of steps before a task is learned. However, due to physical limitations and resource constraints, such extensive training is impractical for real-world applications, especially for the execution of dexterous manipulation tasks. Additionally, training in simulation and successfully transferring the model to real-world scenarios remains a significant challenge due to the sim-to-real gap. This paper addresses these challenges by introducing a Phased Behavior-Guiding Reward Function to enhance the efficiency of RL in learning and executing complex, real-world dexterous, in-hand manipulation tasks. The proposed approach uses evolving rewards across distinct phases, allowing the agent to learn incrementally and produces a cohesive policy for the entire task. Experiments conducted on a low-cost, physical robotic manipulation testbed demonstrate that the phased reward function achieves faster learning and higher task success rates, outperforming single-phased and traditional distance-based reward functions. These results suggest that the proposed method could be a sample-efficient and scalable solution for complex real-world robotic applications. © 2024 Australasian robotics and automation Association. All rights reserved.

关键词： Robot applications

来源：评论

学校读者我要写书评

暂无评论

Development of a 3D, High-Fidelity Simulator for Autonomous Racing of F1Tenth Cars

Development of a 3D, High-Fidelity Simulator for Autonomous ...

引用

2024 Australasian Conference on robotics and automation, ACRA 2024

作者： Steiner, Emily Pama, Afereti Zhou, Futian Lee, Kyla Liarokapis, Minas Williams, Henry Centre for Automation and Robotic Engineering Science The University of Auckland New Zealand New Dexterity research group The University of Auckland New Zealand

Autonomous racing is a new field of research stemming from recent developments in autonomous vehicles. Autonomous racing competitions, such as F1Tenth, provide researchers with opportunities to push the limits of autonomous capabilities. However, it is often impossible or unsuitable to develop and test autonomous racecars solely in the real world, due to time constraints, track availability, and crash risks. Therefore, simulators provide platforms for low-risk, low-cost development and testing. Simulators are often custom-built for a given racing competition, vehicle style, or track environment. Currently, the F1Tenth competition lacks a high-fidelity 3D simulator which allows for seamless transfer between simulation and reality. This research developed such a simulator for the F1Tenth competition. The developed simulator was then used for training end-to-end reinforcement learning algorithms on F1Tenth vehicles. The successful deployment of these algorithms in simulation racing and real racing demonstrates the usefulness and high fidelity of the built simulator. © 2024 Australasian robotics and automation Association. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Benchmarking Reinforcement Learning Methods for Dexterous robotic Manipulation with a Three-Fingered Gripper

Benchmarking Reinforcement Learning Methods for Dexterous Ro...

引用

2024 Australasian Conference on robotics and automation, ACRA 2024

作者： Cutler, Elizabeth Xing, Yuning Cui, Tony Zhou, Brendan van Rijnsoever, Koen Hart, Ben Valencia, David Ong, Lee Violet C. Gee, Trevor Liarokapis, Minas Williams, Henry Centre for Automation and Robotic Engineering Science The University of Auckland New Zealand New Dexterity research group The University of Auckland New Zealand

Reinforcement Learning (RL) training is predominantly conducted in cost-effective and controlled simulation environments. However, the transfer of these trained models to real-world tasks often presents unavoidable challenges. This research explores the direct training of RL algorithms in controlled yet realistic real-world settings for the execution of dexterous manipulation. The benchmarking results of three RL algorithms trained on intricate in-hand manipulation tasks within practical, real-world contexts are presented. Our study not only demonstrates the practicality of RL training in authentic real-world scenarios, facilitating direct real-world applications, but also provides insights into the associated challenges and considerations. Additionally, our experiences with the employed experimental methods are shared, with the aim of empowering and engaging fellow researchers and practitioners in this dynamic field of robotics. © 2024 Australasian robotics and automation Association. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

On Model-Free Deep Reinforcement Learning for Dexterous robotic Manipulation: Benchmarks, Analyses, Challenges, and Implementation Tips

On Model-Free Deep Reinforcement Learning for Dexterous Robo...

引用

2023 Australasian Conference on robotics and automation, ACRA 2023

作者： Valencia, David Jia, John Li, Raymond Hayashi, Alex Cutler, Elizabeth Gee, Trevor Liarokapis, Minas MacDonald, Bruce Williams, Henry Centre for Automation and Robotic Engineering Science The University of Auckland New Zealand New Dexterity Research Group The University of Auckland New Zealand

Applications using Model-Free Reinforcement Learning (MFRL) have grown exponentially and have shown remarkable results in the last decade. The application of MFRL to robots shows significant promise for its capability to solve complex control problems, at least virtually or in simulation. Due to the practical challenges of training in a real-world environment, there is limited work bridging the gap to real physical robots. This article benchmarks the state-of-the-art MFRL algorithms training on an open-source robotic manipulation testbed consisting of a fully actuated, 4-Degrees of Freedom (DoF), two-fingered robot gripper to understand the limitations and challenges involved in real-world applications. Experimental analysis using two different state-space representations is presented to understand their impact on executing a dexterous manipulation task. The source code, the CAD files of the robotic manipulation testbed, and a handy guide on how to approach MFRL’s application to real-world are provided to facilitate replication of the results and further experimentation by other research groups. © 2023 Australasian robotics and automation Association. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Does ChatGPT and Whisper Make Humanoid Robots More Relatable?

arXiv

引用

arXiv 2024年

作者： Chen, Xiaohui Luo, Katherine Gee, Trevor Nejati, Mahla Centre for Automation and Robotic Engineering Science The University of Auckland New Zealand

Humanoid robots are designed to be relatable to humans for applications such as customer support and helpdesk services. However, many such systems, including Softbank’s Pepper, fall short because they fail to communicate effectively with humans. The advent of Large Language Models (LLMs) shows the potential to solve the communication barrier for humanoid robotics. This paper outlines the comparison of different Automatic Speech Recognition (ASR) APIs, the integration of Whisper ASR and ChatGPT with the Pepper robot and the evaluation of the system (Pepper-GPT) tested by 15 human users. The comparison result shows that, compared to the Google ASR and Google Cloud ASR, the Whisper ASR performed best as its average Word Error Rate (1.716%) and processing time (2.639 s) are both the lowest. The participants’ usability investigations show that 60% of the participants thought the performance of the Pepper-GPT was"excellent", while the rest rated this system as"good" in the subsequent experiments. It is proved that while some problems still need to be overcome, such as the robot’s multilingual ability and facial tracking capacity, users generally responded positively to the system, feeling like talking to an actual human. Copyright © 2024, The Authors. All rights reserved.

关键词： Anthropomorphic robots

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：