I've built a lot of robots over the years and one question that's often asked is, “Why don't you make the robot solar-powered?”
My latest robot, Tertill, is solar powered, my previous robot, the HV-100 from Harvest Automation is not. Why? It all boils down to how much power a robot can collect from the sun and how much power the robot needs to do its job. Let's compare the HV-100 (aka Harvey) and Tertill.
By the time sunlight reaches the earth’s surface it is able to deliver about 1000 watts of power to every square meter--if it's noon time and you're near the equator. But round numbers are convenient so let's pretend that’s the case and start with 1000 watts per square meter.
Harvest’s Harvey robot needs 100 Watts to pick up pots and carry them from one place to another. The robot is just about 0.5 meters (18 inches) in diameter so if we cover the top with a circular a solar panel it would have an area of 0.2 square meters. The power available to the robot would be 1000 W/m^2 * 0.2 m^2 = 200 W. We only need 100 W so we're good, yes? No.
Solar panels are not 100% efficient, affordable ones are closer to 15%. When we add reality to the equation our 200 W becomes 200 * 0.15 = 30 W and the robot doesn't have enough power to do its job. We can’t run on direct solar power but maybe we could let the robot sit in the sun for while and charge its batteries before it begins its work day.
During the summer we might get 10 hours of sunlight or 30 W * 10 hrs = 300 W-hrs of energy. That would let the robot run three hours each day. (Actually less because available power declines when the sun is low in the sky or clouds happen by.)
This strategy of operating the robot at most three hours per day would not make Harvey customers happy. Nursery and greenhouse operators pay around $30,000 for their robots. Working only three hours per day the robots wouldn't come close to earning their keep. Harveys must be able to work at least eight hours each day and during crunch time they have been known to run 24 hours per day. Such is the life of a robot. Thus, Harveys must have an externally chargeable battery that can be swapped out to allow them to keep running. These robots simply can’t collect enough energy on their own to do their job.
Tertill has a different mission and different power requirements.
Tertil’s mission is to whack weeds often enough that they never become larger than sprouts. To accomplish this it need only return to the same spot in the garden every few days. It’s even OK to patrol less often when the clouds roll in because weeds grow more slowly then.
Harvey and Tertill differ in size and weight but it turns out that they could both operate for about the same number of hours each day based on the energy available from the sun. In Harvey’s case that’s not enough time to do its job, in Tertill’s case it is.
The bigger the garden, the more time per day Tertill needs to work to ensure full coverage. At some point, as we make the garden larger, we'll need a second Tertill. But a typical garden is well within the capability of one Tertill.
It was most satisfying to finally discover an application where a robot could do its job using only solar power.
(Physics aside: Although often confused, energy and power are not the same thing. Energy is power times time. You get energy when you collect solar power and store it in a battery. Stored energy can be converted back to power to be used by the robot.)