A harvesting system includes a vertical frame, a plurality of linear robots, a plurality of cameras and a processor. The vertical frame is configured to be positioned opposite a sector to be harvested. The robots are arranged in pairs stacked vertically in the frame, each pair including first and second robots that are configured to move together along a vertical axis, to move independently of one another along a horizontal axis, and have respective first and second robot arms that are configured to approach the sector and harvest fruit. The plurality of cameras is configured to acquire images of the sector. The processor is configured to identify the fruit in the images and control the robots to harvest the fruit.