I have already bought 6 Maxbotix (XL-Maxsonars EZ/AE) range finders. I am trying to integrate them in the Arducopter code but i do not understand which part of the code to modify to add these sensors in a daisy chaining format for obstacle avoidance... any ideas ????
Currently there are none working ones. System that Jose played long ago is too inaccurate for production so it's development is temporary paused. There are some interesting sensors coming and if/when they come, we can look this thing again.
"There are some interesting sensors coming..."
Could you please give references for these future sensors ?
We're looking at using doing sonar-based obstacle avoidance using 5-6 Maxbotix range finders and writing some code for it. Would you mind sharing? It would be great if we didn't have to write it from scratch.
Hi Kal, you may find this paper useful. In this paper, some obstacle avoidance scheme with a few Sharp range finders and a maxbotix sonar was implemented.
Do cite this paper if you find it of any help. Thanks a lot!
What's new on this topic ?
Its has been almost 2 years since the last post on this thread. Are there any updates in this area of intereset?
Not really. There is still some sensor issues that needs to be solved first. Most likely proper obstacle avoidance starts to work when LIDAR systems are getting better. There are several projects going on this area but nothing is released yet. Most likely something will be released during this year.
Lidars? Oh no that sounds more like in 10 years :) Lidars are pretty expensive and the latest area of usage I have heard is self-driving vehicles like Google Car. But if that eve happens it it would be interesting. Then we can use lidars for other porposes too ...
Nope, there are several companies making small lidars and some of them are already starting to work. Ok they are not going to be free for sure. Smallest Lidars are really small. Something like 50mm diameter and 80mm high full 360 degree lidar for multicopter use.
Anyways whole obstacle avoidance will be really difficult, it needs both fast sensors and fast computing to make it working. Currently Lidar seems to be only way to do it if we want fast and accurate detection.