WIP - RbtKpfW III Ausf. A "Schneewolf"

[Lauwenmark]

Edit (September 10th): I finally got a video of the beast walking :)

History

The High-pressure Turtle

After the failure of the Robotkampfwagen II project, the OKL - OberKommando des Lego was left with virtually no heavy tank to replace the obsolete T1-Cs, designed more than 30 years before. Professor Edband Heimser would get them out of that dead-end.

Heimser had previously studied the past campaigns of the empire in the central territories. He got to the conclusion that mud always was one of the biggest issues, severely hampering the advance of the armies for several months each year. He thus worked on legged vehicles that would be better suited for such difficult conditions.

The first prototypes built by the Heimser Waffenfabriek AG made attempts to reuse the new HPV-6 (Hochdruck Pump V-6) compressor designed for the RbtKpfW II project. Heimser wanted to use pneumatics to operate the knee joints of the legs, controlling the upper leg joints with more conventional NXT engines. While theoretically correct, that design choice was actually a mistake. The HPV-6 had been designed to power slow-motion devices with no more than two pistons working at a given time; for Heimser's planned use, the compressor was vastly underpowered, unable to maintain the pressure needed for more than a couple knee movements. Attempts to increase the threshold pressure in the pneumatic circuit weren't successful; when five workers were injured by the explosion of a high-pressure conduit during a test, Heimser decided to scrap the idea and went back to the drawing board. The failed prototype - nicknamed Hochdruckschildkröte (The high-pressure turtle) was dismantled at the Imperial Scrapeyards shortly thereafter.

The following pictures were taken after the electronics were removed, shortly before the turtle was scrapped.

The Snow Wolf

The new prototype used a more conventional "pantograph-like" leg design, giving less degrees of freedom than the HPV-6-based robot, but ensuring better stability and strength. Each pair of legs was powered by its own NXT engine. The front and the back were designed as separate entities, joined together by a turntable part to allow the model to turn. Turning was controlled by a third NXT engine powering coupled linear actuators. A tail gave the machine some supplementary stability. The now traditional Ultrasonic detector was used as the primary long-range scanning device; audio capture was added at the front while a contact sensor was appended to the tail to complement the sensor array.

The first test prototype was presented to the OKL and immediately received strong support. The project was classified as R3X-1 under the name of "Project Snow Wolf" (Schneewolf). Minor design changes were made to the steering system, while a more complex rear gear train was implemented to allow the future connection of an electrically-powered 12std cannon. The initial series got into production shortly thereafter under the designation of Robotkampfwagen III Ausf. A. The RbtKpfW III displayed here is such a "A" revision. The "B" revision was externally identical, upgrades being entirely in the controlling, featuring a brand new control panel designed by Lejos Industria.

Although performing as expected, the RbtKpfW III/A and /B revisions weren't without faults; rushed in production before all mechanical issues could be resolved, they lacked several planned features. In the initial design specs, the head could rotate to allow a 135° sonar scan in front of the machine; the /A and /B revisions had a fixed head, forcing the driver to stop and flex the robot body left and right to get a rough equivalent. Moreover, it lacked a complete gear train to power the planned 12std cannon, forcing the OKL to equip it with much smaller artillery pieces. Finally, its rotation radius was quite big, often forcing the driver to perform several back-and-forth maneuvers to make a 90° turn when space was limited. Despite those shortcomings, the RbtKpfW III performed well on the battlefield.

****

Design notes

Nice points: It can move by itself, which is a huge improvement over my previous model ;). It also does what I wanted to get: a remote-controlled walker.

Not so nice points: I wanted to implement a system to derive a new function from the motors - the idea was to use a differential as a "substractor" between the front and back leg motors - the derivation would turn only when both motors were running in opposite direction. I didn't manage that - I was not able to find a way to fit that on the chassis. It would have been fun to use it as a rocket launcher platform, though :). The gear train visible at the rear is part of that - it currently serves no other purpose than moving air around :).

Short-term plansFinish writing the controlling software - I wrote a small Java control panel to command it from my laptop, but it is quite ugly and doesn't really do anything with sensors yet. And who knows, maybe I'll find a way to finally fit that damn 12std cannon on its back? :)

****

Thoughts? Ideas? Opinions for improvement welcome!

Edited September 10, 2010 by Lauwenmark

[efferman]

the "Schneewolf" looks nice. have you a video of it?

[Lauwenmark]

the "Schneewolf" looks nice. have you a video of it?

Not yet - it was too dark already to take one. I'll try to make one Tomorrow morning :).

[Lauwenmark]

Not yet - it was too dark already to take one. I'll try to make one Tomorrow morning :).

*cough* - my camera was non-functional, so the video was delayed somewhat - apologizes for the lateness :)