UTM and Remote ID

Example 1 Four operators file intents for the same morning

OP-A is our pipeline inspection, OP-B a delivery, OP-C a police mission that the authority ranks higher, and OP-D a mapping job. Each is a box in local coordinates, height and time after 09:00 (hypothetical data).

# 4D operational intents: x, y box (m), height (m) and time (minutes after 09:00)
intents = {
    "OP-A survey":   dict(x=(0, 800), y=(0, 600), z=(0, 90), t=(0, 40), prio=0),
    "OP-B delivery": dict(x=(700, 2000), y=(500, 700), z=(60, 120), t=(30, 45), prio=0),
    "OP-C police":   dict(x=(-500, 300), y=(400, 900), z=(0, 120), t=(35, 60), prio=1),
    "OP-D mapping":  dict(x=(1500, 2500), y=(0, 400), z=(0, 90), t=(10, 50), prio=0),
}

def overlap(a, b):
    return all(max(a[k][0], b[k][0]) < min(a[k][1], b[k][1]) for k in "xyzt")

names = list(intents)
for i, a in enumerate(names):
    for b in names[i + 1:]:
        A, B = intents[a], intents[b]
        if overlap(A, B):
            if A["prio"] != B["prio"]:
                who = (a if A["prio"] < B["prio"] else b) + " (lower priority)"
            else:
                who = "same priority: rule set by the authority"
            shared = {k: (max(A[k][0], B[k][0]), min(A[k][1], B[k][1])) for k in "xyzt"}
            print(f"CONFLICT {a} x {b}: time {shared['t']} min, height {shared['z']} m -> replan: {who}")
        else:
            print(f"clear    {a} x {b}")

CONFLICT OP-A survey x OP-B delivery: time (30, 40) min, height (60, 90) m -> replan: same priority: rule set by the authority
CONFLICT OP-A survey x OP-C police: time (35, 40) min, height (0, 90) m -> replan: OP-A survey (lower priority)
clear    OP-A survey x OP-D mapping
clear    OP-B delivery x OP-C police
clear    OP-B delivery x OP-D mapping
clear    OP-C police x OP-D mapping

OP-A conflicts with two operators: with OP-B at 30–40 minutes and 60–90 m, where equal priority means the authority’s rules apply, and with OP-C at 35–40 minutes, where OP-A must replan because its priority is lower. OP-A’s easiest fix is to finish the overlapping part before minute 30 or split the area into smaller parts that take less time. Intents wider than necessary cause more conflicts with others.

Example 2 Will an officer see our drone?

An officer stands beside the pipeline with a broadcast Remote ID receiver. Different environments cause 20%, 50% or 80% of messages to be missed. The drone flies at 20 m/s.

RATE_HZ = 1.0                 # at least 1 message per second (14 CFR 89.310)
LATENCY_S = 1.0               # position broadcast within 1.0 s of measurement
ACC_M = 100 * 0.3048          # position accuracy 100 ft at 95% probability

for loss in (0.2, 0.5, 0.8):  # fraction of messages missed by the receiver (different environments)
    for window in (3, 10):
        n = int(window * RATE_HZ)
        p = 1 - loss ** n
        print(f"loss {loss:.0%}  window {window:>2} s: P(at least one message) = {p:.3f}")

v = 20.0                      # drone speed m/s
worst = v * LATENCY_S + ACC_M
print(f"position uncertainty at {v:.0f} m/s: up to about {worst:.0f} m")

loss 20%  window  3 s: P(at least one message) = 0.992
loss 20%  window 10 s: P(at least one message) = 1.000
loss 50%  window  3 s: P(at least one message) = 0.875
loss 50%  window 10 s: P(at least one message) = 0.999
loss 80%  window  3 s: P(at least one message) = 0.488
loss 80%  window 10 s: P(at least one message) = 0.893
position uncertainty at 20 m/s: up to about 50 m

In the open few messages are missed, and 3 seconds of listening almost guarantees one. But with 80% loss, such as behind trees and buildings, the chance within 3 seconds is below half. The position received may be about 50 m from the true one because of latency plus accuracy. Officers should use this data to identify a drone, not to navigate precisely to it.