A Resource Optimization Approach Based on Deep Integration of Vision-Motion System for Sustainable Agriculture

Abstract

The integration of vision and motion systems represents a critical phase in the intelligent transformation of consumer electronics aimed at enhancing productivity. However, owing to the operational modes and distributional characteristics inherent in existing systems, achieving large-scale, stable, and consistent agricultural applications on consumer electronics remains a significant challenge. To address this problem, this paper proposes a unified resource optimization approach for different configurations of agricultural consumer electronics to achieve deep integration of vision and motion systems. The optimization is mainly at two levels: on the one hand, we design a velocity observer for the vision-motion integrated system represented by the visual servoing system, which makes native changes to the characteristics of the visual servoing’s non-real-time instruction issuance. By converting long-time commands with uncertain period to short-time commands with certain period, the design difficulty of real-time trajectory planning of the real underlying motion controller is simplified. On the other hand, in image-based visual servoing (IBVS) system, the mixture parameter of the image Jacobian matrix also affect the control performance of the visual servoing system. For most IBVS-based agricultural applications, there is a lack of a systematic approach to ensure that the mixture parameter is adaptively and continuously varied. To solve this problem, this paper proposes a fuzzy logic-based method to adaptively adjust this parameter and ensure its continuity by introducing a suitable membership function. The experimental results of visual servoing based on the consumer electronics show that our proposed method can significantly improve the integrated vision-motion controllability, and can trade-off the convergence efficiency and feature retention constraints to effectively improve the overall efficiency of the system operation.