Efficient and accurate object grasping and placement in open hybrid scenes is an important frontier research hotspot in the field of robotics and artificial intelligence in recent years. It has extensive and important application requirements in the fields of industrial manufacturing, home services, etc. Our project takes the intelligent completion of desktop sorting by robots as the typical scene to carry out relevant research. It is proposed to realize the complete perception of objects through independent exploration of the environment, and then reasonably plan the motion trajectory as the research idea. The main contents are as follows: 1. Perceptual recognition and pose estimation of target object combined with autonomous environment exploration; 2. Optimal trajectory planning of manipulator considering object state and collision; 3. Development and demonstration verification of manipulator desktop finishing integrated system.
Efficient and accurate object grasping and placement in open hybrid scenes is an important frontier research hotspot in the field of robotics and artificial intelligence in recent years. It has extensive and important application requirements in the fields of industrial manufacturing, home services, etc. Our project takes the intelligent completion of desktop sorting by robots as the typical scene to carry out relevant research. It is proposed to realize the complete perception of objects through independent exploration of the environment, and then reasonably plan the motion trajectory as the research idea. The main contents are as follows: 1. Perceptual recognition and pose estimation of target object combined with autonomous environment exploration; 2. Optimal trajectory planning of manipulator considering object state and collision; 3. Development and demonstration verification of manipulator desktop finishing integrated system.
MIT License