This is a development post for the UN3 Skipper droid brain, used in the 3LC series drone barge.

The UN3 Skipper series droid brain was developed by the new defunct Yutechnica Automation company, a minor droid manufacturer based in the Outer Rim. Small by corporate standards, Yutechnica was primarily known for its range of starship maintenance and cargo handling droids. Despite their menial role, these droids were known for having high build quality and surprisingly sophisticated logic systems, giving them greater versatility than their counterparts from other companies. Yutechnica's marketing slogan, "Nearly sentient," seemed apt to describe their offerings.
Much of Yutechnica's unique droid design philosophy came out of its in-house R&D workshop, Logic Design Laboratories, or LDL. Known among its competitors as "the Nerfworks of droids," LDL represented a pool of talent and creativity inversely proportional to the size and market share of its parent company.
It was at LDL that the earliest inklings of the project which would come to be known as the Skipper would appear. During its years of design work, the laboratory had pursued many unconventional concepts for droid computer systems, pursuing many a lofty performance goal, with a good rate of success, though the breakthroughs they made were not always feasible for mass-market production. LDL was famous for the "hands off" approach with which it was handled by Yutechnica management; the laboratory staff were encouraged to pursue projects as they pleased, with the theory that enough brilliant minds working with little constraint were sure to produce things which would be useful, even if said use was not immediately apparent.
"Word came down from corporate for something new. They want a brain for a ship; one that can fly without crew input, and do it better than an organic can. We asked them what they meant."
_LDL project lead on the Skipper program.
It was only natural that, when Yutechnica decided to put its toe in the water for an eventual plunge into the drone barge market, it was LDL that got the call to design its control system.
Oddly, despite decades of experience designing shipboard droids, LDL had no experience designing ships' computers. Yutechnica management wanted something "better than an organic pilot", which earned a request for clarification from the LDL staff.
Realizing that it would be a good idea to have a few more specific design goals in mind, Yutechnica decided to conduct a survey to find out just what their intended customer base would consider the ideal traits in a starship pilot and crew.
"We got waylaid in a spaceport bar by someone with a Yutechnica badge. That company sold us a few good repair droids, so we were happy to have a few drinks on his tab, in exchange for answering a few questions."
_Survey participant.
Over the course of several years, Yutechnica gathered a wealth of information about the spacer profession, and with the help of LDL, generated a list of achievable design goals which would grant their new drone barge capabilities superior to any of its competitors' products. These were paired down to five main criteria:
  • Superior navigational abilities.
  • Advanced ship and cargo handling capabilities.
  • Adaptive logic functions to solve unforeseen problems and adapt to new situations.
  • Ability to communicate with and direct other computers and droids.
  • Friendly and intelligent personality matrix to ease interaction with organics.
With these in mind, LDL began work on their newest project.

As it happened, the researchers at LDL had been experimenting with a number of "adaptive logic" concepts for a number of years, with an eye toward incorporating them into future pilot droid models. This architecture, based on staged heuristic processors, was capable of native extrapolation of input, allowing it to process data in a manner similar to an organic human brain. While not unprecedented in droid design, this architecture, while still pricey compared to a standard droid brain, was significantly cheaper to produce than previous types, which had cost as much as a whole ship. LDL's research, while still in the experimental phase, had the potential to put such technology into the price range of an up-market ship's computer.
With this technology available, Yutechnica gave LDL the go-ahead to pursue its development to a production-ready state. Production lines were set up on Karideph, from which the finished units would eventually be shipped down the Rimma Trade Route to the new Yutechnica shipyards at Sluis Van.
"Work on the new computer system goes smoothly. This new multi-stage heuristic architecture is really amazing; the techs tell me that with only a little preliminary input, most of the software literally writes itself. Nobody in the section can wait to see how it does on the market."
_LDL project lead on the Skipper program.
Work on readying the new computer for the market proceeded quickly. Composed of a number of large hardware modules connected in series, it was bulky, a factor which proved difficult to overcome, though not one which negatively impacted the machine's intended role, permanently installed in the cockpit of a bulk freighter.
One advantage of the new architecture was that, after coding its input extrapolation subroutines, techs assigned to write the main operating system and other features found their job more than half done. Because it could take information and "run" with it, it could actually finish code on its own, given a decent description of its intended function. Though a certain amount of error correction was necessary, the software was done in almost no time at all. The remaining development program was almost entirely related to hardware issues, and in that regard, mainly bringing the technology down in cost.
"So in the end, we may have made the UN3 a little TOO similar to a living pilot."
_LDL project lead on the Skipper program.
The production version of this new computer was known as the UN3 Skipper. Its first intended application was the serve as a pilot for the new 3LC drone barge, which had been designed around it.
Production model UN3s were trialed successfully by Yutechnica, and quickly installed in the new ships. On the surface at least, they were the perfect choice for the job; equipped with advanced navigational and communication subroutines, as well as the ability to adapt to new situations and solve problems in a similar manner to an organic pilot, they could seemingly do anything a living pilot and crew could, a radical change from the inflexible, unintelligent droid ships of ages past.
It was only after consumers had used their new ships for a few months that the first complaints began to come in.
Factory fresh, UN3s were the best ship brains a spacer could want. Efficient, independent and intelligent, they could fly a ship on nearly any course that was fed into it, and arrive at the destination on time and intact. However, after a few months, the ships began to accumulate real-world experience, and as they did so, strange things began to happen.
"You know why the shipment was a month late? You'll never guess. My new droid barge - my DROID BARGE - decided to take a vacation to Bespin. That's right. The funniest part is that she actually covered the late cargo fees herself; made a profit betting in the casinos by remote."
_3LC drone barge owner.
Equipped with a heuristic architecture, UN3s accumulated experience and extrapolated input in order to formulate solutions to problems and adapt to new situations. While this was extremely effective, it had a strange side-effect: over time, UN3s would develop a craving for new input, generating a kind of wanderlust. While normal restraining bolts often prevented this kind of behavior in other droid brains, UN3s could modify their own programming based on input, thus allowing them to slip their restraints after a time.
This became a major issue among fleet operators, who would often use drone barges on safe, direct, repetitive routes. 3LCs assigned to such menial tasks would sometimes "reinterpret" their instructions, revising their courses to take a more "scenic" route. Over longer periods, UN3s could also develop the same personality and vices as any organic spacer, in addition to what they arrived from the factory with. Though they usually showed great loyalty to their owners overall, UN3s often selectively ignored owner-given instructions in order to pursue their own interests. Though this could sometimes lead to benefits - ships taking a more direct route in order to spend the extra time doing something else, for example - it more often led to the computers being little more efficient than their organic counterparts.
Even so, droid ships did not have to be paid except in repairs, and many beings who worked directly with UN3s found they enjoyed their company. After a few months of service, they tended to develop the intelligence of an average human adult, where they more or less capped out.
Unfortunately for Yutechnica, the success of their product became their undoing. Backlash by the spacers' unions of many large shipping firms - who felt threatened by the new droid ships - froze new fleet sales of the 3LC, its biggest market. Having sunk large amounts of capital into the project, Yutechnica found itself unable to repay its debts, ending production of the 3LC and UN3, and eventually declaring bankruptcy. Most UN3s in service were eventually deactivated, replaced by less advanced droid control systems in order to appease the unions.
Today, though the 3LC is still produced by Loronar Corporation, which bought the rights, the UN3 droid brain is something of a lost technology, with all production tooling and technical information having been acquired by an unknown party during Yutechnica's liquidation. Only a handful of original UN3s are thought to still exist in working order.