English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

As a kind of intelligent agricultural equipment, picking robots are of great significance for improving the efficiency of agricultural production. However, the main bottleneck restricting the development of picking robots today is the positioning and control in image recognition. Therefore, an agricultural picking control method based on visual servo technology is proposed. This method can accurately control the picking hand of the picking robot on the basis of building an eye-hand relationship model and an online identification system. With the tomato picking process as the background, the effectiveness of the method was verified. The test results of image feature points show that there is a small error between the stable feature points and the expected image feature points. In addition, the image plane trajectory of the picking robot is relatively smooth, and there is no vibration or overshoot. In addition, from its density distribution characteristics, it can be seen that the picking hand movement is a continuous acceleration stage at the beginning of the control, and at the end of the control, the characteristic points gradually tend to the desired characteristics, and the output of the control system is relatively stable. It can be seen from this that the model has better control performance. This method has certain practical significance for the improvement of agricultural picking efficiency. 摘要 采摘机器人作为一种智能农业设备,对提高农业生产效率具有重要意义。然而,目前制约采摘机器人发展 的主要瓶颈是图像识别中的定位与控制。为此,提出了一种基于视觉伺服技术的农业采摘控制方法。该方法在 建立眼-手关系模型和在线辨识系统的基础上,能够准确地控制采摘机器人的拣手。以番茄采摘过程为背景,验 证了该方法的有效性。图像特征点测试结果表明,采摘机器人的稳定特征点与期望的图像特征点之间仅存在误 差。另外,拾取机器人的像面轨迹比较平滑,没有振动和过冲现象。另外,从其密度分布特征可以看出,采摘 手的运动在控制开始时是一个连续的加速阶段,在控制结束时,特征点逐渐趋向于期望的特性,控制系统的输 出相对稳定。由此可见,该模型具有较好的控制性能。该方法对提高农业采摘效率具有一定的现实意义。 INTRODUCTION Agriculture industry is an important guarantee for food security in all countries in the world. Agricultural robots are widely used in agricultural production to improve the ability of agricultural robots to sense their work and precise control, which can not only improve agricultural modernization level, but also improve international status in agricultural production and technology exchange. Among them, the fruit and vegetable picking robot, as a key field of agricultural science and technology research, is of great significance to agricultural modernization. However, today's fruit and vegetable picking robots usually have poor picking results due to positioning errors and motion errors (Bao, Cai and Qi, 2016). In order to overcome the problem of low control accuracy of fruit and vegetable picking operations in complex environments, the article explores the control error of fruit and vegetable picking robots, and proposes a micro-control method of fruit and vegetable picking robots based on visual servo technology to effectively improve agricultural picking efficiency. The micro-control of the picking robot achieves precise positioning in a small space by installing a monocular camera on the robot arm to adjust the image error (Bargoti et al, 2017). Its core lies in setting the control purpose of the robot, so as to automatically obtain image information and analyse it, and output the feedback information of the machine control in a short time to achieve a good effect on the control of the robot trajectory (Mehta et al, 2016). 1 Yujie Jin, As. PhD. Eng. Vol. 62, No. 3 / 2020 INMATEH – 192 According to the observable features on the plane and the texture target plane, a suitable control law is derived, and the estimated factors of unknown constant force disturbance are introduced into the control law, and then the effectiveness of the controller is proved through experiments. (Wang S. et al., 2018). In order to study the visual servo control of space-based space robots, Dong G. and his colleagues conducted an indepth discussion on non-cooperative target pose and motion estimation algorithms. Finally, experiments prove the applicability of the control scheme (Dong et al., 2015). In order to effectively solve the problem of visual servo control for robots without degrees of freedom, the Tsai C.Y. research team proposed a new hybrid switched reactive visual servo control structure, which is derived from perception and reaction behaviour and does not require inverse interaction matrix calculations (Tsai et al., 2015). This control structure helps to simplify the realization process of the visual servo system and improve the control efficiency (Serra et al., 2016). Szymon Krupínski et al. proposed a visual servo control method for underwater robots based on nonlinear inertial auxiliary images. This method uses the homomorphic matrix between two images of a plane scene as feedback information to control the system in a cascaded manner in the control design. The development of dynamics is more robust to the uncertainty and perturbation of the model than the traditional solution that only considers the traditional solution of system kinematics. (Szymon Krupínski et al., 2017). The research team of Silveira G discussed the problem of vision-based robot control and proposed a new strengthbased non-metric visual servo technology. This technology does not consider the target characteristics of the observation object, the displacement of the camera and the relative attitude, and can achieve the dynamic decoupling of the translation control error. Finally, through corresponding experiments, the effectiveness of the technology is proved. (Silveira et al., 2020). Combined with experiments to verify the effector, the results show that the effector has great potential in promoting kiwifruit harvest. The research team outlined the research and development activities of agricultural robots in recent years, and discussed some specific issues related to the precise control of weeds and fertilizer applications by robots. (Wang J, Zhou G and Wei X., 2019). With the development of picking robots, Tao Y and his colleagues proposed an apple automatic recognition method based on point cloud data in order to improve the robot's recognition ability and perception ability in three-dimensional space. (Tao Y., Zhou J., 2017). The research team introduced an intelligent machine that helps farmers pick lemons. The machine is a mechanical arm that can cut lemons from trees. This machine can help farmers complete their work effortlessly, greatly reducing labour (Yan H.J., Bai G. and He J.Q., 2010). In order to manage the precise application of fertilization and water in agriculture, Zavala-Yoe R et al. constructed an agricultural intelligent production model. The original model is equipped with a laser sensor, which can measure the content of water, fertilizer or any other nutrients in the crops. This is of great significance for improving the level of agricultural modernization. (Zavala-Yoe R. et al., 2017). MATERIALS AND METHODS Eye-hand relationship model building The essence of robot visual servoing is the problem of nonlinear system control, which is generally described by the eye-hand relationship model. (Qingqing S., Bin W., Jiawei Y., Yizhi L. and Tongming Y., 2020) Therefore, the first task of constructing an agricultural picking robot model based on visual servo technology should be to establish an eye-hand relationship model, and then use an online identification system to set its real-time status, and finally establish a control strategy design plan, so the agricultural based on visual servo technology is obtained. (Changbin H, Yong Y, Decheng W, Hongjian W. and Bingnan Y., 2020). According to the positional relationship between the camera and the mechanical structure, the eye-hand configuration of the agricultural picking robot can be divided into an eye-on-hand configuration and an eye-fixed configuration. The two configurations are shown in fig.1. The eye-fixed configuration refers to fixing the camera on a certain working scene of the agricultural picking system to ensure that the entire picking process is captured and it is convenient to integrate the image information into the visual control system. The image obtained by the eye-fixed configuration is clearer and the image resolution is higher. (Min J, Chunguang W, and Pengpeng W., 2020) The eye-on-hand configuration refers to fixing the camera on the moving arm of the agricultural picking robot so that the camera and the moving arm are synchronized. The main advantage is that the posture of the robot arm can be flexibly adjusted to minimize the distance between the camera and the target object, so as to ensure the accuracy of picking the target, and at the same time improve the image observation accuracy. Because the eye-on-hand configuration is more conducive to the precise positioning of the target picking object, this time the eye-onVol. 62, No. 3 / 2020 INMATEH – 193 hand configuration is mainly used. In the process of building the relationship model, the most important thing is to define the coordinate relationship.

RECOGNITION TECHNOLOGY OF AGRICULTURAL PICKING ROBOT BASED ON IMAGE DETECTION TECHNOLOGY