How to use Machine Learning image classification to get a drone to fly to an object? (ESP32S3)

[DainyBoi]

Hello all, I have created an ML model of a certain type of radish using Edge Impulse studio, and uploaded it as a sketch on Arduino ide as static_buffer, allowing the esp32s3 camera to determine if the thing it presented with is a radish or not. I have a drone made with esp32s3, a small one with small motors soldered to motor driver boards in a quadcopter-like configuration. How do I use my ML model to get my drone to fly to the radishes? I did some research on things like openCV but I'm struggling to find any leads on how to achieve this. The drone needs to classify an object as a radish, which it already does, and then fly towards it. Do I need a distance sensor to achieve this or can I go based on just a camera? how do I control the speed of the motors on the driver to slow once it approaches? Does anybody have any potential leads or resources on this?