机器人模型接口、sEMG 信号处理与图像分割实验。研究条目按当前阶段介绍任务、方法和评估问题。Work on robot-model interfaces, sEMG signal processing and image-segmentation experiments, presented through the tasks, methods and evaluation questions at their current stage.
作为 SmartCam 的在研扩展，整理视觉与语言输入、模型推理和机器人动作接口。当前工作包括 SmolVLA、Octo-Small 等模型的本地实验，以及感知—决策—执行链路的实验方法。An ongoing SmartCam extension studying visual/language input, inference and robot-action interfaces. Current work includes local SmolVLA and Octo-Small experiments and methods for evaluating the perception–decision–action pipeline.
sEMG 手势识别与适应性研究：整理信号处理、实验配置和评估流程，分析个体差异对模型表现的影响。当前按研究实验阶段开展。Research on sEMG gesture recognition and adaptation, including signal processing, experiment configuration and evaluation, with individual variation as a central question.
通过图像分割模型复现整理数据处理、训练配置与结果评估流程，比较复现条件与模型输出。该条目属于研究方法实践。A model-reproduction study of image segmentation, covering preprocessing, training configuration and result evaluation, with attention to how experimental conditions affect outputs.