My company wants to start integrating AI and autonomy into most of the products we sell. We make almond harvesting equipment and want to develop a fully autonomous fleet. GPS will take care of most the navigation, but there is still many things to sense in order to get the job done. Some examples of this might be following a row of nuts within a row where GPS signal cuts out. I’d use a machine learning algorithm for this, paired with a camera. Id most likely want to send a signal to the IFM controller/display I would be using telling it to turn left, right, or continue straight. This means I might need to send a PWM signal out of the Nvidia device. I want the chip to be able to handle at least a few different algorithms at the same time with multiple output signals. On a separate machine, we may also may want to detect fruit, the coordinates of the fruit using a point cloud camera, detect people and detect fruit bins. We would prefer to have one chip that would work as a platform for all of our machines. Would you have a product recommendation for this ?
Please find the Jetson produce specification from Jetson Modules | NVIDIA Developer to know the module computing capability for you to choice, also can refer those projects sharing at Latest Jetson & Embedded Systems/Jetson Projects topics - NVIDIA Developer Forums to gain some ideas…