I'm reading this survey on the state-of-the-art techniques for indoor positioning and navigation. I noticed that LIDAR isn't mentioned at all in the paper. Is LIDAR not meant to be a common method for indoor navigation? I believe I've seen some industrial LIDAR robots some time ago though.
$\begingroup$
$\endgroup$
7
-
$\begingroup$ The article is mostly about positioning in indoor settings. LIDAR by itself gives no position; and using SLAM will give you a position relative to a perceived map, but not an absolute position. $\endgroup$– JuanchoCommented Aug 7, 2019 at 22:42
-
$\begingroup$ @Juancho But don't many techniques reviewed also need perceived maps? For instance, Bluetooth and WLAN methods need perceived signal strength maps. $\endgroup$– John M.Commented Aug 8, 2019 at 4:06
-
1$\begingroup$ Lidar isn't used for positioning, it's used for navigation. $\endgroup$– tuskiomiCommented Aug 8, 2019 at 13:10
-
1$\begingroup$ @tuskiomi Yes, I believe that paper covers both. $\endgroup$– John M.Commented Aug 8, 2019 at 14:14
-
$\begingroup$ 2D LiDAR or LRF sensors are common for indoor navigation (e.g. Hokuyo URG or Hokuyo UTM) $\endgroup$– Benyamin JafariCommented Aug 8, 2019 at 15:13
|
Show 2 more comments
2 Answers
$\begingroup$
$\endgroup$
1
I think LiDAR is common for indoor navigation. Definitely, LiDAR is the easiest and accurate solution for indoor navigation or SLAM. Many commercial robot vacuums are already in use of LiDAR for indoor navigation and mapping. Those are even cheaper and simpler than RGBD modules which is why low-cost LiDARs are hired over RGBD in mass production models.
It is just that indoor navigation is a vast topic. I haven't seen any survey paper that covers all types of sensors. We can't cover few thousands of papers in a single survey paper. The paper you linked has some specific coverage and it is ok not to mention other sensors out of the scope such as LiDAR or camera.
-
$\begingroup$ It does cover visual SLAM, I believe. $\endgroup$– John M.Commented Aug 8, 2019 at 15:08
$\begingroup$
$\endgroup$
The referenced paper appears to be talking mainly about "mobile devices" (i.e. phones), not mobile robots. I believe there are simply physical limitations to using something like LIDAR on a phone. When holding a small device in your hand and up against your head, most of the device is covered, so optical technologies won't work well. That is why the majority of techniques the paper investigates are radio technologies. (Radio waves can travel around and through human bodies and walls as opposed to line of sight optical sensors like LIDAR).
I'd contend that LIDAR for indoor mobile robots is used extensively. Robots that use 2D LIDAR for indoor SLAM can be found throughout the literature. Including some notable books like Probabilistic Robotics. For 2D LIDAR to be most effective, it needs to be held at a consistent height above the ground. Something that comes naturally to wheeled indoor robots.
The costs of 2D LIDAR have greatly been reduced over the years. Now even commodity vacuuming robots use it.
--> 
--> 
However, I will also mention that in recent years cheap 3D sensors like the Kinect and its derivatives have shifted robotics (and SLAM research) to 3D.
