The goal of this project is to combine a solar power bank and a robot chassis with some obstacle avoidance and irradiance detection capabilities. The robot will be programmed to perform a random search and will use some heuristics to determine optimal solar charging locations.

There are a number of challenges that will need to be accounted for. The largest, and most obvious is that the cheap solar power banks that I am using usually don't have large enough solar panels to effectively charge the attached battery, even in good sunlight. And considering that I am planning on using the power bank battery to power the robot, and that my apartment where the robot will be operating is not particularly sunny, it is probably safe to assume that the overall production of the system will be net negative. My hope is that if I can keep the robot power consumption low enough I can get to positive overall power production, but my hunch is that if I ever plan on using the power bank for what it was originally intended for (charging phones/devices), I will need a much bigger solar panel to ensure enough power for both the operation of the robot as well as charging the batteries.

Another big challenge will be coming up with a good search algorithm. There are only a few good windows in my apartment, and they are spread out between different rooms. Ideally the robot will find it's way to the best location for specific times of the day, but a random search is unlikely to result in achieving that. And adding to this complication is the fact that I will be using hobbyist level obstacle avoidance, which means there is a high likelihood that the robot fails to even navigate a single room without getting stuck.

On the off chance that I get this thing to produce some net watt-hours, I might try and add some remote power monitoring capabilities and possibly a panel pan/tilt system to help get a little more juice form the panels.