Motion capture system for apple picking robot

With the structural adjustment of China's agricultural industrialization and the Internet + The rise of models , Fruit planting plays an important role in generating income for farmers . Compared with the fully automated planting and harvesting of major food crops （ Like wheat 、 corn ）, At present, apple 、 Pear and other fruits with thin skin and easy to damage still need to be picked manually , This greatly increases labor costs , And climbing is very dangerous . Therefore, the development of mechanized apple harvesting robot has important practical significance .

The apple harvesting robot mainly consists of a vision system and a manipulator with an end effector , Covers machine vision 、 The image processing 、 Robot kinematics 、 Multi sensor fusion and other technologies . Although domestic and foreign researchers have developed a number of apple harvesting robots , But there are still many problems to be solved , For example, there is a lack of assessment of the degree of pulp damage 、 branch - Fruit stalk - The modeling method of fruit and the mechanism of fruit separation are not well understood .

In order to reduce the damage of apple pulp during picking , Researchers from Northwest University of agriculture and forestry science and technology have carried out research on the optimization of the picking posture of the apple harvesting robot . First, the researchers set up branches 、 Fruit stalk 、 Finite element model of fruit , And compare Abaqus Simulation results and test results are verified , It provides a theoretical basis for the grasp and damage assessment of the end effector and the optimization of the picking action .

The picking action of the mechanical arm directly affects the fruit picking effect , In order to make the picking effect of apple harvesting robot close to that of man , Used by researchers NOKOV Measuring motion capture systems , The upper arm moves when picking apples manually , Then control the manipulator to realize the picking action according to its track path .

Layout around the site 8 individual NOKOV Measure motion capture camera , In testing a person's shoulder joint 、 elbow joint 、 Reflective identification points are pasted on wrist joints and fingertips , utilize NOKOV The measurement motion capture system tracks and records the trajectory data of people picking apples , And select fingertip trajectory data for research and analysis , For trajectory planning of manipulator

 

 

reference ：

[1] Bolingxin . Research on Key Technologies of apple harvesting robot in structured orchard [D]. Northwest Agricultural and Forestry University of science and technology ,2021.DOI:10.27409/d.cnki.gxbnu.2021.000038.