Further (scathing) Analysis re. Final Report on Police Drone - Cessna Collision

[PBDawg]

On January 21st, I posted a thread on the the report from the Transporation Safety Board of Canada regarding the 2021 collision of a police drone with a Cessna 172 over the final approach to Buttonville Airport. I included a link to a video by Don Joyce with his take on the report. More recently, Don reached out to the TSB, Transport Canada, the York Regional Police, and the owner of the Cessna.

Here's the link to Don's new video with results of his investigations and his scathing indictment of the report:

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

 

[Robert Prior]

Interesting that Transport Canada lists contributing factors "usually in chronological order", and yet the failure of the visual observer to actually observe and the task saturation of the RPAS pilot are listed after the failure of the Cessna pilot to spot the drone.

If they were actually listed in chronological order, the lack of briefing/training for the visual observer and task saturation would come before anything about the Cessna pilot.

 

[Ralph thompson]

The review by this guy is detailed, persistant and broad, kudos for that but he demonstrates (as in his other videos) a lack of knowledge of the world of aviation. For many reasons, legal being the obvious one, the TSB does not assess blame, it’s not their job. They review the facts and present them clinically in order not to show any bias. Cause might be obvious to some (like the guilt of OJ Simpson) but it’s up to a court of law to determine guilt. Not sure what plane was involved here, but the approach speed for a Cessna 172 is 61 knots. The student was under the hood so the instructor had to be 100% vigilant since it’s likely they were flying VFR. The Matrice is a big drone.
however, I’m not defending the police action which does seem incompetent at best.

 

[Robert Prior]

Not sure what plane was involved here, but the approach speed for a Cessna 172 is 61 knots.

From page 7 of the report:

The RPA was in a stationary hover for more than 2 minutes when, at 1301, a collision occurred with the Cessna, which was on final approach for Runway 15 (Figure 1), approximately 1.2 NM from the threshold, and travelling at approximately 65 knots.

Police RPAS operations commenced at 1232, without informing ATC. Plane was a Cessna 172N.

The body of the report (the history section) presents the events in chronological order. The summary, and the headline on the TSB website, is not in chronological order, listing the failure of the Cessna pilot to spot and avoid the drone first. What puzzles me (engineer, non-aviation) is that if the order is not chronological I would assume it to be by significance, which also doesn't seem to be the case. So why did an experienced investigator not follow normal procedure and list factors in chronological order?

I have not read other incident reports to determine how often the causal factors are listed in chronological order and how often they aren't, so I am taking the TSB at their word when they say "usually". Maybe it's 51% of the time? That would actually be an interesting exercise — I'll try to find time this week.

 

[Zbip57]

he demonstrates (as in his other videos) a lack of knowledge of the world of aviation.

I think what's more important is his knowledge of the world of drone regulations. The main point of his video is to question why the drone pilot got off so easily, fined only Cdn$1000 as an individual, rather than the larger fine for a corporation, and for only a single offence that would surely have been dealt with much more severely if it had involved a member of the general public, rather than the police.

For many reasons, legal being the obvious one, the TSB does not assess blame, it’s not their job. They review the facts and present them clinically in order not to show any bias. Cause might be obvious to some (like the guilt of OJ Simpson) but it’s up to a court of law to determine guilt.

Correct, the Transportation Safety Board does the investigation and publishes the report without assessing blame. It is Transport Canada who writes the regulations and enforces them with appropriate fines assessed. The fine was never contested in court.

Neither the student pilot nor instructor were fined for their failure to "see-and-avoid" the collision. That makes perfect sense. What makes no sense at all is that their see-and-avoid failure was listed as the very first item in the TSB's report as to cause and contributing factors leading to the collision.

As Don points out, much larger fines have been imposed for much less severe situations. All of the regulations are intended to avoid presenting a hazard to manned aviation or to people and property on the ground. And here we have an actual collision caused by multiple infractions over several regulations, the result of which could only have been worse if it actually killed someone. And yet the penalty was relatively insignificant.

The student was under the hood so the instructor had to be 100% vigilant...

What makes you think the student was "under the hood" during final approach to landing?

Here's the TSB report:
www.tsb.gc.ca/eng/rapports-reports/aviation/2021/a21o0069/a21o0069.pdf

 

[Ralph thompson]

I think what's more important is his knowledge of the world of drone regulations. The main point of his video is to question why the drone pilot got off so easily, fined only Cdn$1000 as an individual, rather than the larger fine for a corporation, and for only a single offence that would surely have been dealt with much more severely if it had involved a member of the general public, rather than the police.


Correct, the Transportation Safety Board does the investigation and publishes the report without assessing blame. It is Transport Canada who writes the regulations and enforces them with appropriate fines assessed. The fine was never contested in court.

Neither the student pilot nor instructor were fined for their failure to "see-and-avoid" the collision. That makes perfect sense. What makes no sense at all is that their see-and-avoid failure was listed as the very first item in the TSB's report as to cause and contributing factors leading to the collision.

As Don points out, much larger fines have been imposed for much less severe situations. All of the regulations are intended to avoid presenting a hazard to manned aviation or to people and property on the ground. And here we have an actual collision caused by multiple infractions over several regulations, the result of which could only have been worse if it actually killed someone. And yet the penalty was relatively insignificant.


What makes you think the student was "under the hood" during final approach to landing?

Here's the TSB report:
www.tsb.gc.ca/eng/rapports-reports/aviation/2021/a21o0069/a21o0069.pdf

In my recollection of this incident, the left hand seat was a student pilot practicing IFR approaches with a instructor in the right hand seat.

 

[Zbip57]

In my recollection of this incident, the left hand seat was a student pilot practicing IFR approaches with a instructor in the right hand seat.

Interesting. If true, I suspect the TSB report would have mentioned that.

Here's the view from a Cessna 172 on approach to Buttonville. The collision happened about twice farther out than this. Image from the Wikipedia page for Buttonville.

 

[Capt KO]

As a previous 172 owner/pilot (son 172 owner/instructor now) it’s hard enough to see and avoid other small planes, much less see a drone. A matrice against the ground clutter, nearly impossible. Lessons learned..hopefully. Tragedy avoided.

 

[Robert Prior]

A matrice against the ground clutter, nearly impossible. Lessons learned..hopefully. Tragedy avoided.

I used Apple Maps to grab a view of what the airport looks like from just before the collision location in this post:

Final Report on 2021 Police Drone Collision with Cessna - A Very Interesting Analysis

Is it usual to mention things like that in investigations into collisions? Yes. If this was an incident report in my own field (education), mentioning that someone didn't see something would imply that they should have seen it. And listing it first in the findings would imply that it was the...

mavicpilots.com

 

[Robert Prior]

As a previous 172 owner/pilot

Quick question for you. Assuming you could spot the drone while landing (over a built-up area, if that makes a difference), how long would you need to safely see it, evaluate it, and avoid it?

Is there a minimum distance for seeing things over the nose of the Cessna? Like, if it was closer than xx metres it would be under the nose? I'm assuming the plane would be a bit nose-up in relation to the flight path while descending, but that's based on watching Mentour Pilot's simulations of jets landing so not a strong assumption.


I'm putting together an assignment for physics students, and I'd like to have realistic numbers not only for physics stuff (how far can our eyes resolve an object, how many metres per second the plane travels, etc) but also for the piloting stuff (how far away must the pilot recognize the drone for what it is to safely avoid it).

OK, maybe not such a quick question.

 

[Capt KO]

Quick question for you. Assuming you could spot the drone while landing (over a built-up area, if that makes a difference), how long would you need to safely see it, evaluate it, and avoid it?

Is there a minimum distance for seeing things over the nose of the Cessna? Like, if it was closer than xx metres it would be under the nose? I'm assuming the plane would be a bit nose-up in relation to the flight path while descending, but that's based on watching Mentour Pilot's simulations of jets landing so not a strong assumption.


I'm putting together an assignment for physics students, and I'd like to have realistic numbers not only for physics stuff (how far can our eyes resolve an object, how many metres per second the plane travels, etc) but also for the piloting stuff (how far away must the pilot recognize the drone for what it is to safely avoid it).

OK, maybe not such a quick question.

Pilots reaction time varies greatly. At least with planes and birds you immediately know the direction of travel but a drone..who knows. The time it takes to figure out what you're looking at, it could be too late. Depending on your forward speed it could be a hundred yards or a quarter mile, just too many variables. If someone yells, Think Fast, often the opposite happens.

 

[Robert Prior]

Pilots reaction time varies greatly. At least with planes and birds you immediately know the direction of travel but a drone..who knows. The time it takes to figure out what you're looking at, it could be too late. Depending on your forward speed it could be a hundred yards or a quarter mile, just too many variables. If someone yells, Think Fast, often the opposite happens.

We know the speed of the Cessna (65 knots), so it's an easy physics problem to convert between distance and time.

What I'm not certain about is the reaction time. In a car where the decision is simple — brake! — it still takes at least 0.7 s to react even when the driver is scanning the road in front of them. (Much longer if distracted.) Once the car starts braking it is a simple deceleration problem. (Swerving would also be an option, but in traffic that's likely to cause more problems than it solves, and also people are less likely to do it when potentially hitting something smaller than their vehicle.)

So I can easily make up problem sets involving cars and pedestrians/dogs/etc in the road, providing some real-world context to "why would we want to calculate that?". Also helps that I've been driving for half a century.

As an experienced pilot, suppose you suddenly realize that there is a drone hovering right in your flight path while you are landing. (Put aside for the moment the question of how long it would take to see the drone and process this information*.) What maneuver would you take to avoid the drone? Climb, turn, both? (I'm assuming not dive in this situation.) How much time/distance would it take for the plane to make the maneuver, once you start it? If the drone isn't moving you don't have to move more than half the wingspan laterally to avoid it (with no margin of error!), so avoiding the drone by turning looks like a simple question of turning arc but I can't find information on how long it takes to start turning once you begin moving the controls.

All of this is a gross simplification, I know, just as modelling the stopping of a car as (constant velocity based on reaction time) + (stopping distance based on coefficient of friction) is**. But it does give students an idea of the physics behind the situation.

(Not that a pilot would be calculating this, any more than an athlete calculates trajectories when they kick/throw/hit a ball.)

*I'm having lunch with a cognitive psychologist later this week. I'm planning on asking him for some pointers to research on reaction/processing times. I'm certain someone's done it.

**Come to that, 2H2 + O2 ==> 2H2O is also a simplification, and a pretty big one, but it's a useful simplification.

 

[zinnware]

Here is a Flight Simulation of the Cessna 172N flight on August 10, 2021 that collided with a a drone near the Buttonville Municipal Airport outside of Toronto, Canada. It included approximate time of day weather conditions and collision location as reported by the Canada Transportation Safety Board in their report. See if you as a Cessna pilot could pick out the stationary drone.

To view this content we will need your consent to set third party cookies.
For more detailed information, see our cookies page.

Accept third party cookies

 

[Robert Prior]

Here is a Flight Simulation of the Cessna 172N flight on August 10, 2021 that collided with a a drone near the Buttonville Municipal Airport outside of Toronto, Canada.

The map on figure 1 of the report (on page 7) shows the collision happening just to the west of Leslie Street, roughly between Cassandra Crescent and Love Court.