类器官芯片装配数据集

器官芯片技术构建功能类器官实验主要是建立肝脏、胰腺或心脏等人体器官芯片的新技术体系，实现功能类器官的体外构建。然而，实现这一目标的关键步骤是类器官芯片的装配，因为它直接决定了器官功能。为了提高实验的准确性和无菌化，用机器人实现器官芯片的装配是器官芯片技术构建功能类器官实验中不可或缺的一步，可以实现复杂器官的体外制造。目前，常见的器官芯片构建通常通过实验人员进行，耗时耗力，但由于实验人员手部的力量控制不精确，以及在多次试验下实验的无菌性无法满足实验的条件。本文通过基于一种具有模块化和高度可定制化的机器人控制系统，搭建了面向类器官芯片装配的控制系统。
首先，完成了轻型低惯量机械臂设计工作。本文通过对于需要实现类器官芯片操作的技术指标进行分析，完成机器人低惯量关节的设计和高刚度连杆的计算，通过对于机器人负载能力的计算，完成了机械臂关节长度的确认。通过机器人的正运动学与逆运动学分析，完成了从机械臂关节角与机械臂末端位姿的对应关系，并完成了相应的仿真验证，基于解析法完成了机械臂的工作空间的分析与计算。
其次，详细阐述了通过视觉信息与力觉信息完成引导矢量的构建。通过DenseFusion的方法完成了的RGB数据信息和点云信息进行整合。通过机械臂末端提供的六维力信息，完成了基于力觉感知的装配过程分析，提出了一种改进型螺旋线策略，完成类器官芯片的装配任务。通过基于扩展卡尔曼滤波的方式，完成了两种引导矢量的构建，完成了基于视-力信息融合的引导矢量构建任务。
然后，基于引导矢量的类器官芯片装配的柔顺操作任务，完成了基于类器官装配任务中的笛卡尔空间轨迹规划任务，并对于直线规划和圆弧规划进行仿真分析。对于机器人对于微小系统较难操作，对于基于位置的阻抗控制算法进行仿真实验，研究了在不同的阻抗参数下对于系统稳定性的影响，通过仿真实验分析了阻抗参数对于装配性能的影响。
最后，搭建基于面向类器官芯片装配的硬件系统，通过基于提供了一种具有模块化和高度可定制化的机器人控制系统和机器人的人机交互软件系统将视觉检测系统与类器官芯片进行连接，完成机器人实时通信的任务。经过详细的研究，我们利用引导矢量的在线轨迹生成技术以及位置的阻抗控制技术，成功地完成了类器官芯片轴孔的装配，这一结果不仅显示出我们的研究结果的准确性，也为我们的系统提供了可

The experiments of constructing functional organoids using organ-on-a-chip technology mainly aim to establish a new technical system for human organ-on-a-chip devices such as liver, pancreas or heart, so as to realize the in vitro construction of functional organoids. However, the key step to achieve this goal is the assembly of organ-on-a-chip, as it directly determines the organ function. To improve the accuracy and sterility of experiments, using robots to realize the assembly of organ-on-a-chip is an indispensable step in the experiments of constructing functional organoids via organ-on-a-chip technology, which can realize the in vitro manufacturing of complex organs. At present, common organ-on-a-chip construction is usually performed by experimenters, which is time-consuming and labor-intensive. Moreover, the accuracy of force control of experimenters' hands is insufficient, and the sterility of the experiment cannot meet the experimental requirements after multiple trials. In this paper, a control system for organoid chip assembly is built based on a modular and highly customizable robotic control system.

First, the design of a lightweight low-inertia robotic arm is completed. This paper analyzes the technical indicators required for organoid chip operation, completes the design of the robot's low-inertia joints and the calculation of high-stiffness links, and confirms the length of the robotic arm joints through the calculation of the robot's load capacity. Through the forward kinematics and inverse kinematics analysis of the robot, the corresponding relationship between the joint angles of the robotic arm and the end-effector pose is established, and the corresponding simulation verification is completed. Based on the analytical method, the analysis and calculation of the robotic arm's workspace are carried out.

Second, the construction of the guidance vector through visual information and force information is elaborated in detail. The RGB data and point cloud information are integrated using the DenseFusion method. Through the six-dimensional force information provided by the end of the robotic arm, the assembly process analysis based on force perception is completed, and an improved spiral strategy is proposed to complete the assembly task of organoid chips. Through the Extended Kalman Filter (EKF), the construction of two guidance vectors is completed, and the task of constructing guidance vectors based on visual-force information fusion is realized.

Then, for the compliant operation task of organoid chip assembly based on the guidance vector, the Cartesian space trajectory planning task in the organoid assembly task is completed, and simulation analysis is carried out for linear trajectory planning and circular trajectory planning. Aiming at the difficulty of operating micro-systems with robots, simulation experiments are carried out for the position-based impedance control algorithm, and the influence of different impedance parameters on system stability is studied. The influence of impedance parameters on assembly performance is analyzed through simulation experiments.

Finally, a hardware system for organoid chip assembly is built. The visual inspection system is connected with the organoid chip through the modular and highly customizable robotic control system and the human-machine interaction software system of the robot, so as to complete the real-time communication task of the robot. Through detailed research, we use the online trajectory generation technology based on guidance vectors and the position-based impedance control technology to successfully complete the shaft-hole assembly of organoid chips. This result not only demonstrates the accuracy of our research results, but also provides a viable