Lead with the design premise, because it is the unusual part: this is propulsion engineered around the assumption of failure. Most thruster design chases efficiency and thrust on the nominal case. Lockheed's filing starts from the question of what happens when a unit dies in flight, and builds the architecture so the mission survives it.
The grant US10913551B1 (inventors Neil Evan Goodzeit and John B. Henderson), classified in B64G 1/401 (spacecraft propulsion) with B64G 1/007 and B64G 1/26, claims a fault-tolerant, scalable, high-thrust arrangement. Scalable and fault-tolerant in the same claim is the tell: thrust is distributed across multiple units so the loss of any one does not strand the spacecraft.
The mechanism is redundancy with controllability. It is not enough to have a spare thruster; the control system has to keep the spacecraft pointable and maneuverable with whatever propulsion remains. The dependent claims are where that graceful-degradation logic lives, and where the real engineering moat sits.
This is a recognizable defense-prime instinct applied to space: mission assurance over peak performance. A commercial smallsat operator might accept a higher failure rate across a large fleet; a prime building a high-value, possibly crewed or strategic asset designs so that single-point failures do not end the mission. The patent encodes that priority.
What the grant cannot tell you is the efficiency cost of the redundancy, or how often the failure mode it guards against actually occurs in flight. It tells you Lockheed considered fault tolerance worth a high-thrust architecture and a patent. The flight record, classified or commercial, is where the tradeoff gets settled.