It is highly unlikely that uncrewed cargo aircraft will be allowed to operate in most US airspace any time soon. They can't reliably see and avoid other aircraft operating under VFR, and so are restricted to only limited designated airspace.

Airships are unable to cruise at high altitude due to loss of lift, and are vulnerable to damage from severe weather. For ocean routes it's not always possible to route around storms.

People keep wanting cargo airships to be a thing for some reason. It's not likely to happen. The costs are too high and the range of potential applications too limited to produce a real industry. At most we might see some limited military use where cost is less of a factor.

If/when forecasts will be good enough to only dispatch and route airships when they can be safely flown to either destination or safe harbor, is the wildcard I could imagine to make them viable options. But as you say, is the niche large enough to make work?

I think the pull for wanting them isn't so strange - they offer the promise of much lower fuel costs, which is a big stigma and problem of current aircraft.

The atmosphere is a chaotic system. How could forecasts be improved enough to enable safe flights across the Pacific Ocean during storm season?

Concerns over fuel costs seem a bit silly as those are only a fraction of air cargo costs. There are significant fuel efficiency improvements already in the development pipeline with lighter composite structures, higher aspect ratio wings, open rotor turbine engines, and perhaps even blended wing-body fuselages.

Are remote crews an option?

Not an option under current FAA rules. The available optical sensors are still generally inferior to human eyes in terms of dynamic range, depth perception, and slew rate. The US military does fly remote-piloted aircraft (Predators being the most prominent example) but they're only allowed to operate in limited pieces of designated airspace due to the risk of midair collisions.

Communications reliability and latency is a problem. We still have no way to guarantee solid bidirectional comms. The mishap rate for RPVs is much higher than for comparable manned aircraft.

I'm not comfortable relying on these being crewless, I've seen too many mobility projects die. When freight trains and trucks can operate crewed, a cargo airship will have no different rules to profitability.

I don't think you can send perishable goods via a very slow and delay prone transport medium. More likely this can be used for the opposite sort of goods: durable, low urgency supplies.

They are only 'very slow' when compared to aircraft, which are kind of poor vehicles for transporting cargo in the first place.

A dirigible flying with the jetstream is almost twice as fast as a cargo ship doing the same.

I think they're impractical for lots and lots of other reasons, and your "delay-prone" critique is probably salient, but "slow" needs to be contextualized somewhat.

Twice as fast as a cargo ship is still only 50kph right?

And the Jet stream only goes one way and only West to East (in the northern hemisphere) and only at certain latitudes right?

So if you're only competing against cargo ships. And you happen to want to head East (only, no returns). And you need to go faster than a ship, but not over 100kph. And you are already at the right latitude and so it your destination. And you're cargo is not going to perish any time soon, and is not too dense, then this can work?

You're exactly right, but perhaps missed that the trade route you're describing is China to Los Angeles, which alone accounts for $132 billion in trade every year, and perhaps that cargo vessels allegedly account for 20-25% of anthropomorphic carbon emissions.

Also probably worth pointing out that airships going against the jetstream are still faster than cargo ships which are also going against sea currents.

I'll repeat my disclaimer again here, that "I think they're impractical for lots and lots of other reasons" but there is a definite benefit to cutting the carbon emissions of the world's most popular trade route by 90% and halving the time spent in transit even if you assume that there are no other applications, which is probably not correct.

I also find the idea of "they have to land at a airport" pretty murderous for autonomous vehicles. Like no they do not. They can deliver cargo to autonomous barges, which then deliver the cargo slow and steady and reliable to cargo terminals. No autonomous vehicle ever has to touch land..

100%. It seems obvious when thought about (but most people haven't thought about it) -- but airships don't need runways. Moreover, the only land-space that needs to be accommodated for them is the footprint, so a landing spot could be on the top of a skyscraper within a city center, or a parking lot, stadium, etc.

And that's not even counting the possibilities for water landings