Jidong Lv

Apple picking robot obstacle avoidance based on the improved artificial potential field method

On the basis of retaining the merits of the traditional artificial potential field method which has simple structure and is easy to implement, and combining with the characteristics of obstacles in the apple growing environment, an improved artificial potential field method is present for the obstacle avoidance of apple picking robot. The virtual target point is introduced to help the searching progress escape from local optimal minima for some existing shortcomings, such as local minima, the stuck district. The productivity of apple picking robot is improved effectively. The experimental result proves the feasibility of this method.

Research on the Strategy of Advancing Harvest Efficiency of Fruit Harvest Robot in the Oscillation Conditions

Take the fruit harvest robot experimental platform as background to research on the strategy of advancing harvest efficiency. Sample continuous and real-time fruit image by image processing, preprocess to detect the objective fruit, and then create an image coordinate system. Calculate oscillation frequency of the objective fruit with curve fitting of sample value. At the same time, get the depth of the objective fruit with visual and calculate the movement time of the Direct-acing joint to achieve the control of the end effector. The research will reduce the locating time, advance harvest efficiency of the harvest robot when spherical fruits like apple are oscillating.

Dynamic Regulation of the Weights of Request Based on the Kalman Filter and an Artificial Neural Network

Magnetic and inertial measurement units (MIMU) are currently being explored as a promising tool for attitude tracking of a moving object, such as human body parts. The function of attitude calculation is realized by using attitude algorithms. The overall performance of these algorithms is seriously influenced by linear acceleration of the moving object. Therefore, there is a requirement to find solutions to this problem. In this paper, a new attitude algorithm for MIMU known as REQUEST is introduced. The algorithm is then revised in order to be suitable for MIMU. A new representation of linear acceleration of the moving object is then constructed. An artificial neural network (ANN) is used to establish the functional relationship between this representation and the weights assigned to the vector measurements in REQUEST, for the purpose of adaptively regulating the vector measurements according to the representation. In this way, the measurements of the gyroscope can be relied on more for attitude calculation when the carrier has a higher linear acceleration. A Kalman filter (KF) is also used prior to the establishment of the functional relationship in order to take full advantage of historical sensor information for accurate estimation of the representation. Our experiments have verified good static and dynamic performances of our KF+ANN-based REQUEST algorithm.

Research on trunk and branch recognition method of apple harvesting robot

In order to provide fairly perfect environment information for automatic navigation and harvesting obstacle avoidance of apple harvesting robot, the trunk and branch of apple harvesting robot was studied. The most effective multiple segmentation method of trunk and branch was selected after the discussing and comparison, then the morphological closed operation and holes filling operation were executed, the central axis that is the main characteristics of trunk and branch that was extracted through skeletonizing method and removing subbranch method based on different templates. Finally, the tests about recognition effect and recognition time for trunk and branch were done, the detecting rate was 95%, and the recognition average time was less than Is.

Peach fruit recognition method under natural environment

Peach fruit has inhomogeneous colors including green, white and red. The work aimed at studying peach fruit recognition method under natural environment. Firstly, peach was conducted with image segmentation pretreatment using OTSU dynamic threshold segmentation method based on R-G color description. Fruit region was demarcated in original image. Secondly, peach fruit was distinguished using improved NNPROD algorithm after accelerated optimization. At last, the effectiveness of the method was evaluated by experiments.

Yellow Apple Recognition Method under Natural Environment

Recognition method of yellow apple fruit under natural environment was studied in the work. Firstly, image segmentation was conducted using K-Means clustering based on normalized R+G-B color feature. Meanwhile, the work adopted algorithms of Flood Fill for hole filling and elimination method of regional threshold for noise elimination. Then, edge detection of segmented images was implemented based on Canny operator, improving randomized Hough transform method for application in fruit recognition of apple image. Finally, experiments were carried out on recognition of apple fruit, indicating feasibility and effectiveness of the proposed method.

Design and Research on Vision System of Apple Harvesting Robot

The vision system of apple harvesting robot was researched and designed to make it possible for realizing automatic harvesting of apple. The vision system model is studied. The vision system of apple harvesting robot was designed by two aspects, including hardware composition and soft architecture. The VFW method was employed to realize the real-time image acquisition. The recognition of vision system is developed using the combination method of regional growth algorithm and color characteristics. The preliminary orientation of apple target was calculated by finding its centroid. At last, the performance of this version system was evaluated. The results showed that the developed vision system of apple harvesting robot successfully achieved the recognition and orientation of apple target. It was conclude that this version system was effective for realizing the automatic harvesting of apple.

Acquisition of fruit region in green apple image based on the combination of segmented regions

To acquire the fruit region of green apple image in close color more completely, this work designed a method of region extraction before fruit region combination. Firstly, channel images in every color space were analyzed to determine appropriate color channel images: images of G and b color channel in RGB and Lab color spaces, respectively. Then, these images were segmented by K-means clustering to obtain segmented images of G and b color channel. Post-processing was conducted to different segmentation results. For example, open operation and micro-region elimination were conducted to segmented image of G color channel. The same processing was conducted to b color channel image after combining with above processed G color channel image and filling holes. Therefore, the target fruit region can be acquired completely. The validity of this method was verified by experiment.

Research on branches recognition of apple harvesting robot

Branches recognition of apple harvesting robot were studied to provide rich environment information for robots harvesting obstacle avoidance. Firstly, the leaves and other background in the apple image were removed based on R-G color difference. Secondly, image segmentation was conducted using K-Means clustering based on “a” color feature in the Lab color space. Then the noise removal and image perfection operation were executed, and the central axis - the main characteristics of branches - was extracted through skeletonizing method and removing subbranch method. Finally, experiments were carried out on recognition of apple branches, indicating feasibility and effectiveness of the proposed method.

Application research of neural networks in fruit and vegetable harvesting robot

Neural networks was used in the camera calibration of binocular stereo vision, segmentation and recognition of fruit or vegetable images, fruit or vegetable classification, orientation error amendment of fruit or vegetable, obstacle avoidance of robot manipulator, sensor data fusion about fruit and vegetable harvesting robot. The research results in all the aspects mentioned above of fruit and vegetable harvesting robot with neural networks were introduced in this article, and some the highlighted issues were finally discussed. It is expected that this research may be used as a step stone for further study in the area.