Single-sided continuous optoelectrowetting (SCOEW) is a versatile light-controlled digital microfluidic technology. However, the absence of automated algorithms for generating light patterns to manipulate droplets has constrained SCOEW chips to employing only pre-recorded sequences. This limitation hinders SCOEW platforms in effectively addressing common experimental challenges like droplet pinning. To overcome this issue, we have developed a set of algorithms for the automatic generation of droplet-driving light patterns. By integrating these algorithms with machine vision (using YOLOv3) to detect, locate, and classify droplets through video camera input, and employing a path-planning algorithm (A*) to optimize droplet trajectories, our system achieves intelligent functionalities. These include simultaneous manipulation of multiple droplets, automated merging based on color-coded droplet composition, obstacle avoidance, and resolution of potential droplet manipulation failures (such as droplet pinning). In a practical application, we implemented an intelligent loop-mediated isothermal amplification (LAMP) assay on the platform. This modular approach to droplet processing is easily adaptable to other SCOEW hardware, offering potential integration of customizable functionality. The result is a promising suite of portable tools for a variety of biochemical analysis application.