Hello, I work in an industry where we have dangerous rooms with hostile environments and we are trying to find ways to reduce the number of manned entries into those rooms. These are rooms with thick concrete walls, lots of equipment and piping, often times very hot (50°C+), and dark. It would benefit us greatly in the planning stages of troubleshooting jobs if we had a way to fly a UAV into these rooms to video the components in question. Because of the nature of this industry, it is vital to be able to avoid bumping components, so a upward viewing camera would be very beneficial. Oh, and GPS is unavailable because of the concrete. Is there something on the consumer market that would work for us? Thank you very much.
It would seem at first glance that a robot which was on the floor would do a better job. There is nothing currently with true obstacle avoidance in the consumer space - or hardly in any other space. Intel just demoed some of that tech at CES but it's likely to be a year or two before this filters down to production models.
Thank you for the reply. We are looking at several Small Unmanned Ground Vehicles as well. SUGV is an excellent choice for many applications, but it has weaknesses that a multirotor would help overcome. In our case, spaces with vertical ladders leading to upper elevations or mezzanines, components in the overhead near the ceiling, or obstacles such as large pipes blocking the pathway that a ground vehicle couldn't overcome. We are looking to round out a robotics program with a diverse toolbox. I'll look at what Intel is doing. Hopefully if enough people like me are expressing a need then they will make sure the product gets to market. I have also been reading about 3DRobotics introducing an indoor flight mode to the next release of their flight control software. Any thoughts on that/them? Thanks again.
3DR and most all other companies tend to over promise and under-deliver. It's not that they don't mean well, just that aerial robotics are much harder than they seem. It took many years to even have a 1/2 way reliably quadcopter for basic video - and every step up will take a long time. Here is the intel demo - as you can see, it has indoor obstacle avoidance. http://gizmodo.com/intel-is-just-daring-autonomous-drones-to-rise-up-and-k-1677888710 But these are demos! As you probably know, it's a long way from there to something that can be sold and supported.
The only good systems I have seen indoors rely on external cameras installed around the flight area, these send signals back to a computer that "reads" the UAV's location / orientation from markings on the UAV and then sends a controls signal back to the UAV to control it. So you don't only have the cost of a UAV, but the added cost of a monitoring system and its complexity to even run it. There are lots of sensors that could be added to a UAV ( ultra sonic, I/R, etc. ) but these tend to be unreliable in a closed environment. Ultra sonic, they can pick up their own echo.....I/R, florescent lights will effect them.... Like everything though, it is being worked on........its just time and resources needed to complete the R and D.
