Since the last post, our rover has undergone several iterations both for the claw and for the body.
We tested the car on an uphill incline but was ultimately unable to climb it. Thus, we changed out the smaller wheels for taller ones. We then doubled the wheels because the weight of the vehicle and outer shell that was later developed made the wheels splay outward. Further changes to the rover body to make the front lower to the ground and make the body bounce less allowed us to remove the previously placed Lego weighed blocks. Thus, the front of the vehicle became more stable and the design used less pieces than before to achieve a new goal.
After this iteration, every part of the rover worked independently: the body drove successfully on flat ground while the claw was able to operate. However, our biggest problem became incorporating the claw in a spot that would allow it to pick up our Mars rocks without a complicated code. Previously, we placed the claw where the Sciborg bumper was. However, that would have required the rover to bypass the Mars rocks and back up to identify the rocks, a largely difficult and complex process for what we were trying to achieve. We considered several different places to place the claw including having it hang off to one side. However, that would have caused an imbalance on one side of the rover. Thus, the only spot that made sense for the claw was to put it at the front of the vehicle. Designing this would have required us to build an extension piece on top of the front-most wheels which would have extended above and over it. This would have created a previously unforeseen problem: since the vehicle was being powered from the back two wheels, the rover would have been unreasonably long and unevenly distributed the weight with two separate heavy bodies attached in the center. The vehicle would not have been able to drive successfully.
This forced us to remove the two front-most wheels to add room for the claw. We were initially concerned that without these two front wheels the robot would not be able to go up or down hills. However, it performed much better than expected especially since the wheels were much taller. But we had to be careful in where we placed the claw and at what angle since it is the front-most piece of the rover. During test runs, the claw sometimes became caught on the ground or posed a problem for when the rover was traveling up and down hills. Several iterations of the claw resulted in a more efficient design and an interesting insight. With our previous code, opening the claw resulted in a 50% success rate because the claw would open fully but then sag back to a half open state. We later realized that if the gears reached 180-170 degrees, it would stay open. But if the gears were less than that, the weight of the claws would naturally cause the gears to rotate and shut the claw. Thus, tweaking the initial starting and ending angles of the claw jaws enabled us to solve this problem.
After several more test drives, the rover still had some difficulty making it up hills. We suspected that the batteries may have played a role in this but was unable to change them until later. Before then, we tested whether powering the vehicle from the front two wheels would solve the problem and were surprised
After several test runs on the landscape's hill, we were finally successful in getting the rover over the majority of the hill. However, because of the claw's weight in the front combined with the motor weight in the middle, the rover would fall on its face when it drove downhill and was unable to rock back to its four wheels. We attached some Lego weighed blocks in the back to try to counter the weight but were disappointed to find out that it didn't help solve the issue as it had done previously. However, we were more surprised to realize that we could simply attach two rods to the front of the claw to push the rover back when it first hit the ground.
Of course there are still more things we would like to iterate upon that we will take the initiative to change between now and the first exhibition.
Looks like models:
We chose to add delrin pieces to the claw to increase its efficiency. The claw features a small curve that was achieved by attaching the pieces via piano wire. These pieces were then hot glued to the claw.
We also created a box to hide the Picoblock board and replicate the look of NASA's rover. Thus, we researched the rover design and spray painted it to model the rover. To replicate the camera head of the rover, we created a stand for Eunice's phone. During the exhibition, we plan on video conferencing the user to the phone on the rover to simulate the experience of viewing Mars from the camera on the rover.
Our landscape went through several design changes. We initially put several rock pieces together in the same area to make the landscape rockier. However, we were concerned that the rover wouldn't be able to make it over all of them if the entire terrain was like that. Thus, we spaced them out more and included some larger rocks for aesthetic purposes. However, when we tested the rover on the actual landscape, it had difficulty making it over some of them due to the ragged edges and required us to shave some of them down.
The hill also took an incredibly long time to create. The angle of the hill required several test drives and us to shave down increments. The first hill I tried to create failed at the end because the last cut made the hill too steep and it became impossible to save it at that point. Thus, I recreated the design using a much smaller piece to begin with and just taped two halves together instead of trying to shave off two sides of one piece. This allowed me to keep the side that worked and create another half when the first didn't work out.
We also created a poster board with instructions on how to operate our rover and included some facts about the Mars rover program to educate our users on the topic.
There are always things to work on that we will definitely consider and implement before our first public exhibition.
