Spiderbot: Struggles Remembered and Suggestions
By Kristine Abatay – Project Manager
Matthew Clegg – Computer & Control Systems
Simon Abatay – 3D Printing & Manufacturing
Following the mass lay-off of all members of the Robot Company for Spring 2014, we remember the numerous struggles faced by Spiderbot with the following video:
Many improvements can be done to make future Spiderbots better machines.
For one thing, connection of Spiderbot to the Arxterra Android App can definitely be helped by possibly finding a forest-like route in a different location on campus. Many issues occurred on demonstration day during the race against Rover and Hexapod. Connection was constantly dropped when multiple projects were connected to the app in the particular race course area and perhaps a location with better wifi connection might remedy that issue.
A route with a more even terrain might be helpful in preventing damage to the gears of the servo motors, as was done with one of the servos from a rear leg of Spiderbot.
A convenient trick we learned from Tien Dang, Communications member of Hexapod, was that if, when taken apart, the servo motor shows gear damage, then the motor can easily be fixed by replacing the gear, which can most likely be purchased online. This is much cheaper than purchasing a whole new servo motor.
The damage was most likely done when testing Spiderbot in the race location prior to the race, when it encountered the various branches of the course.
As for manufacturing, the individual pieces can be improved to help Spiderbot’s overall performance. For one thing, the tibia side portions of Spiderbot’s tibia pieces can be shaved down. The pieces were extended mostly for aesthetics, so if they were shaved down to the same width as the servo motors in their particular orientation, then they should function just the same and be lighter in weight in the long run.
The base of Spiderbot was ultimately very dense, causing the stance of Spiderbot to be higher than it should have been in order to support the entire piece. A wider base most likely would have allowed for a lower position and therefore, a faster sweep of the legs. Research regarding a possible ratio between the base width and leg length should be done to achieve a faster speed.
In regards to the connections of the bracket pieces, the tibia pieces were molded in a manner that made it difficult for the screws to easily tread and lock into them. Because of this, epoxy was initially used, but after testing in the grass, it proved to be too weak of an adhesive. Loctite is recommended as a better adhesive alternative to insure secure connection. That, and testing in terrains aside from grass, since the grass most likely caused some wear and tear in the servo motors from the legs getting stuck.
Purchasing stronger servos would be a good alternative to increase the speed as well, but will ultimately rack up the cost of the final project. Another thing to keep in mind in regards to cost is that creating a mold to cast 3D pieces will require a lot of material, so the initial starting price for this project is already projected to be a couple hundred dollars, so mold and cast wisely.
Good luck to all future Robot Company Projects!