Autonomous outdoor obstacle avoidance using LiDAR

Hi,

So I want to work on obstacle avoidance using a LiDAR in an outdoor environment. Hence I cannot provide initially a map for the environment. I went through ROS and observed that most algorithms require to move the robot around initially to form the map. This however is not possible in outdoor environments. So is there any algorithm which is suited for this task.

Thanks

Asked by Withered_Shadow on 2019-05-30 04:42:39 UTC

Comments

Why is it not possible up move your robot to build up a map because it is outside?

Asked by PeteBlackerThe3rd on 2019-05-30 15:16:15 UTC

Here I assume the obstacles are static but their positions unknown. I would like to give the robot a certain goal and make it move from its current position to the goal, without initializing a map.

Asked by Withered_Shadow on 2019-05-30 22:56:14 UTC

how about using google map. Then use traveling salesman problem (TSP) and have a recalculation every time it encounter an obstacle.

Asked by Usui on 2019-06-06 12:35:39 UTC

Answers

I don't see the problem here. If you want to avoid obstacles, then by definition you are moving. If you are moving then you can build an incremental map. If you mean you need to do global navigation then it simply becomes a matter of replaning your trajectory as you go when you find obstacles. I believe the ROS navigation stack supports this.

Asked by JustinBlack02 on 2019-06-08 18:28:08 UTC