Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Automatic target tracking

Status
Not open for further replies.

kissandra79

New Member
Hi. I am trying to make a device that will automatically track and shoot a moving target using a paintball gun. The target will most likely be a small ball thrown into the air.

I would like to do the entire project in such a way that it incorporates elements of C++, ROS(Robot Operating System), Arduino, Control Systems and Image algorithms for tracking the target. I would like to do it this way.. because I would like some experience using ROS and have fun at the same time.

Here is an overview of how I think that the entire system might work. This is something that I am trying to pull right out of the top of my head. So I could be wrong. Or there might be a better way to do it.

1. A camera acquires an image of the moving target.
2. This information is sent to the computer which is running ROS.
3. ROS does some calculations.
4. Based on the results of the calculations, ROS sends position control information for the paintball gun to an arduino.
5. Arduino controls the motors that turns the turret and pulls the trigger of the paintball gun.

I am from an IT background. So programming and embedded systems are my skills. Im not so good with electronics and mechanical engineering. I realized that the best way for me to go about doing this project is to get the mechanical and electronic components as a kit instead of trying to make them myself. But ofcourse, I intend to write the entire control code, in ROS, and in the arduino...myself.

I have already mentioned the scenario in which the device will be tested. There will be a paintball gun mounted to a pan and tilt system. And this paintball gun will be targeting a moving ball. I am not good with mechanicals as I mentioned earlier. But it is obvious that a pan and tilt system that has to target a moving ball, needs to be a little faster and also be able to move the weight of the paintball gun with ease. And I believe the pan and tilt system will have to shoulder the weight of both the camera and the paintball gun. So I assume that this pan and tilt system must be sturdy and vibration free without any jerky movements. I imagine the ball to be a sphere with radius 2.5 cm.

I searched on google for a pan and tilt system that might be able to do something like this. But most of the information just flys over my head and sometimes I can't seem to find the information that I need.

Could someone please tell me, if there are any places where I can buy a pan and tilt system that can accommodate the requirements of this project? I can afford to spend 250$ if need be...
 
I would suggest spending $25,000 or so on R&D to see if that is even possible. Image interpreting would be seriously difficult, and you are going to need some expensive kit to move a paintball gun in two directions at speed with accuracy.

The cameras should be stationary. You will need several, as you need to know the distance that the ball is, as well as its direction. If you move the camera, it will be very difficult to separate the ball from the background, which would also appear to be moving. Start with some fast servos aiming a laser, and see if you can keep that near the ball before going any further. Your $250 might pay for a couple of servos that would control the laser, by moving a couple of small mirrors.
 
The SPT200 Pan and Tilt System seems like it might suit the purpose of this project. The website says, that it has a maximum payload limit of 2 pounds and I am sure that would not be enough for the requirement of this project. So, I was wondering if it would be possible to change to more powerful servos that can handle more weight and maybe speed? The SPT200 comes without the servos anyways. But what I would like to know is what servo would be more appropriate... The website says i can go for any hitec servo of "Standard" size..

Would the HS-7945TH be appropriate for the requirement?

Thanks
 
You can see that through a linked page from that provided by Jbroadway (https://www.youtube.com/watch?v=kMBojjc6sVg) there's an example of something similar to what you're requesting. The paintball gun is stripped to reduce weight and connected to a couple of RC servos; the remaining mechanics consist of some cardboard and chewing gum.

The camera is fixed (servo control is open-loop); image processing is extremely simple (not sure what was used in above but background subtraction would work).
 
Small revision......

Considering my situation, Instead of using a paintball gun.. I reasoned, what difference would it make if i were to use a laser pointer. Because, using a laser pointer would eliminate many of the nuances that I would have to take into consideration during the project. Ultimately, if my goal is to learn how to use ROS combined with arduino, then it makes sense to go with a laser pointer..

I understand that the project might not be so interesting now after all. But i think it still sounds quite fun..

Apart from significant cost savings... i think this will allow me to concentrate more on the parts that im interested in.

Could someone please tell me.. by how much the difficulty would be reduced, on a scale from 1 to 10, by taking this approach? i am still a little skeptical about going ahead with this.. because I believe I have like maybe 2 months and 10 days to finish this. And the most complicated project ive done so far is controlling the motion of the head of my old barbie doll.. a few months ago..
 
You could try writing object tracking code on a PC using a video of a ball being thrown. You can simulate the laser by painting a red cross on the ball as it travels. You could even include servo inertia and responsiveness into the simulation quite easily. Since this is the hardest part of the problem, it is a good no cost way to test the feasibility.
 
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top