Recognition and Grasping of Disorderly Stacked Wood Planks Using a Local Image Patch and Point Pair Feature Method

Abstract

Considering the difficult problem of robot recognition and grasping in the scenario of disorderly stacked wooden planks, a recognition and positioning method based on local image features and point pair geometric features is proposed here and we define a local patch point pair feature. First, we used self-developed scanning equipment to collect images of wood boards and a robot to drive a RGB-D camera to collect images of disorderly stacked wooden planks. The image patches cut from these images were input to a convolutional autoencoder to train and obtain a local texture feature descriptor that is robust to changes in perspective. Then, the small image patches around the point pairs of the plank model are extracted, and input into the trained encoder to obtain the feature vector of the image patch, combining the point pair geometric feature information to form a feature description code expressing the characteristics of the plank. After that, the robot drives the RGB-D camera to collect the local image patches of the point pairs in the area to be grasped in the scene of the stacked wooden planks, also obtaining the feature description code of the wooden planks to be grasped. Finally, through the process of point pair feature matching, pose voting and clustering, the pose of the plank to be grasped is determined. The robot grasping experiment here shows that both the recognition rate and grasping success rate of planks are high, reaching 95.3% and 93.8%, respectively. Compared with the traditional point pair feature method (PPF) and other methods, the method present here has obvious advantages and can be applied to stacked wood plank grasping environments.

Model Analysis and Experimental Investigation of Soft Pneumatic Manipulator for Fruit Grasping

“…Robotic fruit harvesting has been a research hotspot for over 40 years, as it can tremendously reduce the reliance of fruit growers on a largely seasonal and often untrained labor force [1][2][3]. However, nowadays, most of these robots still use traditional rigid manipulators, which show poor adaptability in grasping fruits of different sizes and shapes [4,5].…”

A Fast Point Cloud Recognition Algorithm Based on Keypoint Pair Feature

“…A method to improve the recognition rate and reduce the calculation time was proposed, but the applicability of this algorithm is relatively simple. Xu [25] proposed a recognition and localization method combining local image blocks and PPF, using deep convolution training images to improve the recognition effect of the algorithm, but it is too cumbersome to implement, and requires a large number of images for training to obtain better results. Bobkov [26] and others also combined convolutional network and PPF, and proposed a 4D descriptor convolutional neural network, which has strong advantages in high noise and occlusion scenes.…”

Active Disturbance Rejection Adaptive Control for Hydraulic Lifting Systems with Valve Dead-Zone

“…Hydraulic lifting systems are applied widely in modern industry owing to their advantages including large force/torque output, small size-to-power ratio, and high response [1][2][3][4][5]. However, considering that heavy nonlinearities (i.e., friction nonlinearity, transmission nonlinearity, and valve dead-zone) and unmodeled uncertainties (i.e., parametric uncertainties and unmodeled disturbances) exist in hydraulic lifting systems, achieving high-performance motion control for hydraulic lifting systems is still challenging [6][7][8][9][10].…”