柔性电极植入机器人的位姿测量和对准插入控制 | |
宋雨佳 | |
2024 | |
页数 | 81 |
学位类型 | 硕士 |
中文摘要 | 脑机接口(Brain-Machine Interface,BMI)旨在建立大脑与计算机之间的通信渠道,实现对人脑活动的实时监测和解读。近年来,柔性电极作为侵入式脑机接口的重要组成部分,因其更好的生物相容性和稳定性受到广泛关注。然而,柔性电极植入手术存在操作难度大的问题,需要借助植入机器人来提高植入精度和效率。本文以柔性电极微创植入为具体应用,重点研究柔性电极植入过程中的识别定位、测量对准、插入控制等关键技术,旨在突破柔性电极植入瓶颈,实现柔性电极沿颅脑微孔自动植入,为脑机接口技术的应用提供技术支持。论文主要内容如下:
|
英文摘要 | Brain-machine interface (BMI) aims to establish a communication channel between the brain and the computer, enabling real-time monitoring and interpretation of human brain activities. In recent years, flexible electrode, as an important component of invasive BMI, has attracted widespread attention due to their superior biocompatibility and stability. However, flexible electrode implantation surgery suffers from the problem of difficult operation, requiring the assistance of implantation robot to improve accuracy and efficiency. This paper takes minimally invasive implantation of flexible electrode as a specific application, and focuses on key technologies such as detection and localization, measurement and alignment, and insertion control during the flexible electrode implantation. The aim is to break through the bottleneck of flexible electrode implantation, realize automatic implantation of flexible electrode along skull hole, and provide technical support for the application of brain-machine interface technology. The main contents of the paper are as follows: (1) Due to the susceptibility of microscopic camera images to factors such as defocusing, occlusion, and lighting, an object detection and keypoint localization network is proposed. In addition, a pyramid structure encoder is introduced for feature map extraction. Furthermore, leveraging the object detection results and pyramid structure feature extractor, the localization of the implanted needle tip and the center point of the flexible electrode ring is realized, which improves the accuracy and robustness of the key point localization. (2) An image feature extraction method based on template matching is proposed for implanted needle and flexible electrode, and an image feature extraction method based on feature point matching is proposed for skull hole. On the basis of image feature extraction, the 3D attitude vector of the target is recovered based on polar constraints and triangulation, and the error analysis of the attitude measurements of implanted needle, flexible electrode and skull hole is realized. In addition, the pose alignment control process is designed and the visual servo control based on the image Jacobian matrix is established to realize the automatic pose alignment of the implanted needle and flexible electrode, and the implanted needle and skull hole. (3) As the implanted needle may collide with the wall of skull hole during insertion, this paper proposes a multimodal contact state recognition network aimed at accurately identifying the relative contact states between the implanted needle and the skull hole. In addition, a control flow for the insertion of the implanted needle through the skull hole is designed. Based on the results of the contact state recognition network, the next moment action of the implanted needle is predicted, and the position servo control is established to adjust the implanted needle attitude, so as to achieve the application goal of smooth implantation of the implanted needle along the skull hole. |
关键词 | 脑机接口 柔性电极植入 位姿测量 对准插入控制 |
语种 | 中文 |
文献类型 | 学位论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/57616 |
专题 | 毕业生_硕士学位论文 |
推荐引用方式 GB/T 7714 | 宋雨佳. 柔性电极植入机器人的位姿测量和对准插入控制[D],2024. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
柔性电极植入机器人的位姿测量和对准插入控(3736KB) | 学位论文 | 限制开放 | CC BY-NC-SA |
个性服务 |
推荐该条目 |
保存到收藏夹 |
查看访问统计 |
导出为Endnote文件 |
谷歌学术 |
谷歌学术中相似的文章 |
[宋雨佳]的文章 |
百度学术 |
百度学术中相似的文章 |
[宋雨佳]的文章 |
必应学术 |
必应学术中相似的文章 |
[宋雨佳]的文章 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论