Research Assistant
Shenzhen Institute of Robotics and AI (AIRS)
03.2024 - Current
- Designed and developed an intelligent equipment system integrating a UR5 robotic arm, iFLYTEK speech module, and an L515 depth camera, capable of performing advanced image segmentation and semantic understanding. Engineered spatial coordinate transformations to align target and tool coordinates for precise object grasping.
- Implemented multimodal interaction through speech-to-text technology, feeding instructions into a large-scale model for layered decision-making. Developed logic for action sequences, controlling the UR5 robotic arm via model output, demonstrating proficiency in human-machine interface design.
- Conducted research on SAM2 and CLIP applications in visual perception, enhancing the system's visual recognition capabilities. This research contributed to improvements in system performance and interaction, showcasing innovation in integrating visual and linguistic data processing.
