Welcome Matrice Pilots!
Join our free DJI Matrice community today!
Sign up

Public Safety - Following a vehicle

Moved to correct section.

How would this be possible? You do not state what country you are in so we have no way of knowing who's aviation laws you need to adhere to but in any case within a couple of minutes the UAV would be BVLOS and would be illegal to continue in most countries
 
Moved to correct section.

How would this be possible? You do not state what country you are in so we have no way of knowing who's aviation laws you need to adhere to but in any case within a couple of minutes the UAV would be BVLOS and would be illegal to continue in most countries
I am in the USA. By achieving maximum altitude permitted and using pan and zoom a vehicle can be "followed" without the aircraft moving. Tests have indicated in the flatlands of Kansas we can track a vehicle up to 5 miles in any direction. In a town it will vary but is reliable in a radius of 2 miles. The objective is to avoid high speeds while tracking and intercepting the vehicle. With an H20 camera and M300 drone a loiter time of 30 minutes is safely achieved. Many vehicle chases end in that time span.

While it might look good in the movies, small UAS aircraft are not practical for chase vehicles in pursuits. In reality small municipalities and rural LE agencies cannot support rapid response helicopters, but with coordination and training sUAS can put another tool in the box and promote public safety by avoiding or reducing the need for high speed vehicle chases.
 
Hmm, so, in order to accomplish that, you’d have to be notified there’s a chase somewhere, get in your car speed to the approximate location, or be guided to where the police are, jump out of your car, unload your drone, start it up, get through the boot up sequence, get up to altitude, then catch the speeding driver on video feed?

I'm guessing the chase would be long over or the driver would be long gone. So unless you just happened to be in the right place at the right time, it ain’t gonna work.

i Just don’t see how it could work even if you got permission. Even if the police had one in their cars, by the time they pulled over, unloaded the drone got through boot up etc, the car would be long gone
 
  • Like
Reactions: licensed pilot
Hmm, so, in order to accomplish that, you’d have to be notified there’s a chase somewhere, get in your car speed to the approximate location, or be guided to where the police are, jump out of your car, unload your drone, start it up, get through the boot up sequence, get up to altitude, then catch the speeding driver on video feed?

I'm guessing the chase would be long over or the driver would be long gone. So unless you just happened to be in the right place at the right time, it ain’t gonna work.

i Just don’t see how it could work even if you got permission. Even if the police had one in their cars, by the time they pulled over, unloaded the drone got through boot up etc, the car would be long gone
Your comments were exactly where my head was at - then in the past 3 days we had 3 pursuits. The most likely for an impact was the rural. Understand that in Kansas roads are generally a checkerboard of squares generally a mile between roads. If a vehicle drops from sight there are limited places for it to emerge and if it doesn't then you are into a search rather than a pursuit. What changed my mind was the stolen property search the following day. At 400 feet I could follow the roads easily while hovering in place overhead out to a radius of 7 miles with the H20 payload. Visibility was perfect that morning. Having pre-staged the previous day in an observational location I would have been able to follow the vehicle to where it got stuck trying to hide down a mud road. The location was visible from the air but not from the travel-able roads. I never launched during the chase because we were all operating under the assumption that drones can't chase cars - and that part is right for the most part.

The evening of the first pursuit a second chase took place in an urban setting. The pursuit lasted for roughly 35 minutes - more than enough time for me to travel to a staging location and launch. Again no launch for the aforementioned reasoning. That pursuit saw the suspect traveling at high rates of speed in a 2 square mile area and covering the same ground several times over. Speeds were excessive for urban interface and pursuit was "backed off" several times for safety. The resources required for that evolution were great. The following day I did a feasibility check to see if I could follow at maximum allowable height AGL. I was able to maintain contact albeit manually - tracking just doesn't work well at those ranges - in three out of four exercises.

In both of these cases I deployed to move-up or pre-stage in the event the suspect bailed and ran on foot or became a barricaded suspect. In essence reducing the response time in those events. In any case permission is not required for SOG response so there is no delay there.

That night a high speed pursuit in a straight line to a crash occurred - there was no chance of observing in that case. Still I pre-deployed in the event forensic mapping was required.

So, to your points, there are several salient points to consider. First is that as a public safety responder and firefighter - my equipment is in ready mode continuously. I am typically in the air well within 2 minutes of arrival on scene - pre-flights are done each morning, batteries are sequentially charged to maintain the ability for continuous flight ops, and the response vehicle is set-up to support rapid and sustained response. In addition, our county is not so large so response times from my central location are within the scope of possibility in many (but not all) cases.

There are a couple of motivating factors. For rural pursuits we really need to know the general direction i.e. roughly west etc. From that I can tell whether or not to get in front of the action or not bother to respond. We are a low population region so there aren't that may LEO's. In a pursuit they strive to maintain contact - not overtake. Any tool in that toolbox is going to have a positive effect. Most of these end up with lost contact, stuck vehicle-dead end road, or TA with the suspect bailing and hiding. In that case we always consider that there will be a search.

Finally, in all cases the safety of life and property is paramount. There are too many variables to list them all but bottom line is lower speeds mean less chance of harm - to everyone. Tracking (not pursuing) with a drone may have a chance to contribute to a favorable outcome. So, we keep looking at our SOG's, pushing the envelope of our equipment, and keep abreast of new technology. To the last point, the Z30 didn't have the reach with resolution and the M200 series didn't have the endurance. The M300 and H20 are changing that. At some point I hope to try the hand-off of flight control feature but for now we'll work with what we have.

In closing I'll add that I've been flying emergencies for over 40 years. My first was the Teton Dam collapse in Idaho in 1976 - That was in a Cessna 150. We never have enough or the right kind of equipment, we never get much advance warning, and the high risk low probability nature of most of these event means the training and budget goes to the high probability events. Still we continue to move forward.
 
  • Like
Reactions: Slaw1211
Sounds like you’ve got a great situation to do that, and that you’ve given proper thought to it. So hopefully it works out. If it even results in catching 1 bad guy, it seems worth it.

The fact that you are a firefighter also helps the situation. Not sure if a civilian could do that.

Good luck and don’t forget to post any good footage here! ?
 
I
Your comments were exactly where my head was at - then in the past 3 days we had 3 pursuits. The most likely for an impact was the rural. Understand that in Kansas roads are generally a checkerboard of squares generally a mile between roads. If a vehicle drops from sight there are limited places for it to emerge and if it doesn't then you are into a search rather than a pursuit. What changed my mind was the stolen property search the following day. At 400 feet I could follow the roads easily while hovering in place overhead out to a radius of 7 miles with the H20 payload. Visibility was perfect that morning. Having pre-staged the previous day in an observational location I would have been able to follow the vehicle to where it got stuck trying to hide down a mud road. The location was visible from the air but not from the travel-able roads. I never launched during the chase because we were all operating under the assumption that drones can't chase cars - and that part is right for the most part.

The evening of the first pursuit a second chase took place in an urban setting. The pursuit lasted for roughly 35 minutes - more than enough time for me to travel to a staging location and launch. Again no launch for the aforementioned reasoning. That pursuit saw the suspect traveling at high rates of speed in a 2 square mile area and covering the same ground several times over. Speeds were excessive for urban interface and pursuit was "backed off" several times for safety. The resources required for that evolution were great. The following day I did a feasibility check to see if I could follow at maximum allowable height AGL. I was able to maintain contact albeit manually - tracking just doesn't work well at those ranges - in three out of four exercises.

In both of these cases I deployed to move-up or pre-stage in the event the suspect bailed and ran on foot or became a barricaded suspect. In essence reducing the response time in those events. In any case permission is not required for SOG response so there is no delay there.

That night a high speed pursuit in a straight line to a crash occurred - there was no chance of observing in that case. Still I pre-deployed in the event forensic mapping was required.

So, to your points, there are several salient points to consider. First is that as a public safety responder and firefighter - my equipment is in ready mode continuously. I am typically in the air well within 2 minutes of arrival on scene - pre-flights are done each morning, batteries are sequentially charged to maintain the ability for continuous flight ops, and the response vehicle is set-up to support rapid and sustained response. In addition, our county is not so large so response times from my central location are within the scope of possibility in many (but not all) cases.

There are a couple of motivating factors. For rural pursuits we really need to know the general direction i.e. roughly west etc. From that I can tell whether or not to get in front of the action or not bother to respond. We are a low population region so there aren't that may LEO's. In a pursuit they strive to maintain contact - not overtake. Any tool in that toolbox is going to have a positive effect. Most of these end up with lost contact, stuck vehicle-dead end road, or TA with the suspect bailing and hiding. In that case we always consider that there will be a search.

Finally, in all cases the safety of life and property is paramount. There are too many variables to list them all but bottom line is lower speeds mean less chance of harm - to everyone. Tracking (not pursuing) with a drone may have a chance to contribute to a favorable outcome. So, we keep looking at our SOG's, pushing the envelope of our equipment, and keep abreast of new technology. To the last point, the Z30 didn't have the reach with resolution and the M200 series didn't have the endurance. The M300 and H20 are changing that. At some point I hope to try the hand-off of flight control feature but for now we'll work with what we have.

In closing I'll add that I've been flying emergencies for over 40 years. My first was the Teton Dam collapse in Idaho in 1976 - That was in a Cessna 150. We never have enough or the right kind of equipment, we never get much advance warning, and the high risk low probability nature of most of these event means the training and budget goes to the high probability events. Still we continue to move forward.
It appears you answered your own question. When in doubt, deploy.
 

Members online

Forum statistics

Threads
2,726
Messages
25,386
Members
5,600
Latest member
RomanChrz