Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

As a member of 341, we competed against 233 in Boston, at Champs, and then again at IRI. It became a game of chicken...who could get to the bridge first, push harder, and push for longer to get those balls. Good times.

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

I voted yes. It provides another facet to the competition which can enhance robot performance. It has the potential to raise the floor, so to speak, and the ceiling.

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

Emphatically NO! I cannot believe that we allowed it this year. The intent of the rules were clear to my team. Autonomous control uses pre-programmed instruction, using sensors to create the changes in robot actions, no driver actions or human input to drive the machines. If we want hybrid control, then it should be called hybrid control. Thanks for bringing this up Karthik.

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

I voted no. The chess matches on Einstein were fun to watch and Cheezy Vision was a great way to work around the hot goal issues this year. However, as a programming mentor for my team I feel that a pure, non-hybrid autonomous presents a better challenge. The teams have to figure out ways for the robot to interact with its environment without human input.

I realize that 254 and 1114's hybrid autos were really just instructions to go left and go right, but this is a slippery slope. What's to stop a team from holding up colored cards to control a wider variety of actions?

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

Playing devil's advocate: Then why not just do away with autonomous altogether? With a little more programming, Kinect/Chezy Vizhun can completely replace the joystick so why not just let the teams use their joysticks?

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

This is why its going to go away next year I believe. Back in 2012 we tested the kinect for driving the robot and preforming actions (there was even a team who balanced a bridge in build with the kinect), and with a bit of practice we were able to do most stuff we would need for auton. Due to wanting a more reliable auton, and a variety of robot problems we never ended up using it. In games where you can play defense in auton a defender with good programming can shut down all but the best teams it seems.

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

A lot of people don't realize that Team 254's 2012 Hybrid Mode was basically all teleoperation with the Kinect.

Evidence: www.youtube.com/watch?v=ZaOiaC0I8pY

/s

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

I voted no. I actually thought about indicating the hot goal using the kinect when I first started thinking about how to detect it, but I determined that autonomous meant autonomous, and this method would most likely be against the rules. I never checked the Q&A, but if I had, I don't think I would have used the kinect anyway.

Moral of the story: autonomous should be autonomous, nothing else.

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

I voted no, but I believe the game makers need to define one way or another if this is allowed in autonomous mode. It was magical to watch the finals and I think it can add a lot of strategy to a 'pre-teleop' period. I just don't think it is correct to say it is autonomous mode if drivers are able to manipulate the robot.

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

Now, how do people's opinions change if the autonomous mode is not at the beginning of the match?

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

I will reiterate that the finals on Einstein likely would have been the EXACT SAME sans kinect and cheesey vision.

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

I love this question. Not that it changes whether a robot is a "Robot" or just a big "RC car", but it introduces a new dynamic to the game, and one I have been hoping would be played out one of theses coming seasons.
Think about how moving Autonomous to the middle or end of a match would change both offensive and defensive strategies!

Agreed!
When you get to that level of performance, there will always be solid ways of addressing the challenges at hand.

Personally, although I like the ability to use things like Kinect and CheeseyVision, I prefer to leave Autonomous, fully autonomous.

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

What is the difference between a Kinect unit or a webcam located on the drivers' station reacting to input and influencing the robot's action as compared to a camera or other type of sensor which is mounted directly on the robot and reacting to input? One difference is how they each are counted (or not counted) in the weight limit ... that could be an argument for excluding their use.

Is it purely an autonomous mode or is it a hybrid or is it almost indistinguishable from teleoperated? How is the programming challenge that different? Why does that matter?

We used a Kinect to initiate the autonomous routine to launch the ball and then drive forward. It enable us to detect a Hot goal and hit it consistently. We employed a simple solution using available technology to meet a challenge.

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

We postulated that Autonomous would be at the end of the match this year, given some pre-season Dr. Who references. It would be an interesting twist to autonomous, but would also make it much harder to accomplish due to greater variance in starting conditions.

Generally speaking, I agree with Adam - some compromise is probably the best way forward. Either way, it will have to be addressed - non-wired robot interaction (i.e. cameras/etc) will definitely increase as time moves on an technology gets better or simpler ideas get introduced. The video game market has come up with some really cool immersive ideas lately.

Re: The use of the Kinect and Cheesy Vision in 2015 and beyond
 

I think the biggest difference is who is processing data to make a differential decision. For "indirect" controls, the driver is making those differentiations, whereas in autonomous, those must be made by the robot according to the programming. Don't get me wrong, I think that the Kinect and CV controls are interesting and fun, and I credit teams like the Cheesy Poofs for being creative enough to develop them within the rules, but a true autonomous should be just that - autonomous.