I'm not saying anything for or against this technology, but very large companies spend a huge amount of money getting hundreds of patents every year. Not so much because they want to build all this stuff, but because they want a huge portfolio of patents so that if "they" get sued for patent infringement, they can find something they patented that the "other guy" has violated. Sort of a Mutually Assured Destruction approach to intellectual property litigation. What counts is not really the quality of the patent portfolio, but the sheer size and the gargantuan breadth of the coverage of patents.
It's a silly thing to have to do, since most of the patents are frankly dumb, obvious, unoriginal, farfetched or impractical. But large corporations are forced to adopt this strategy since courts will happily punish companies with billions of dollars of damages. You've heard of getting "all lawyered up"? Well corporations also have to be "all patented up".
So you can budget for quite a large number of dubious patents being awarded in any given year. Because of the length of time involved, usually the patents are going to be concerned with the technology that was considered to be absolutely, blazing hot about 2-3 years before the patent is actually issued ... hence, drones in theme parks!
It's not correct to say courts are "punishing" companies with billions of dollars of damages. There have only been 5 or 6 patent infringement damage awards that have hit the $1 billion figure, and there are none more than $2 billion. Most if not all were awarded by juries, not judges, and most were later reduced or vacated by judges. In fact, half of the top 25 patent damages awards ever were subsequently reduced or vacated. That's in large part because damage awards must be tied to the value of the use of the patent - damages are not punitive, they are compensatory for the use of the invention (although punitive damages can be levied for egregious infringement or conduct).
Also, while it's true companies spend money on patenting the fruits of their R&D, including technology that they may never actually use, and applying for patents is in part for offensive and defensive patent litigation purposes, patenting their technology is not the primary goal nor really that substantial of an additional expense.
Disney and other companies aren't investing millions of dollars in R&D to create new patents as it sounds like you're suggesting - they're investing that money into developing new technology for use in their business. Patents are secondary and used primarily to protect those investments. And for large companies with big R&D operations, the additional cost of patenting technology after all the R&D investment is really not all that significant, so it makes sense to go ahead and do it (1) if it is something you want to use for a competitive advantage, you can exclude others, and (2) for the defensive purposes you described.
Also, note that by saying "most of the patents are frankly dumb, obvious, unoriginal, farfetched or impractical," you are saying the USPTO improperly issued most patents. Certainly, there are many patents that should not have been issued, but under U.S. law, an invention may not be patented if it is obvious, unoriginal, farfetched, or impractical, if by impractical you mean it it doesn't work as intended or is not useful (if you mean impractical for business/economic reasons, that's not required by law).