Thanks for that definition, ScotCan. If that's the accepted definition of "fully autonomous," we're definitely not there yet in robotics. I agree, HMI plus partially autonomous robots makes a lot more sense.
It depends on the interpretation of autonomous. My understanding is that fully autonomous means a robot's capacity to learn from its environment and carry out its actions accordingly.The human factor processes much more information than any computer because of the wide ranging human response to its environment...most times to the benefit of circumstances, but sometimes in error. That's why HMI makes more sense than attempting to build fully autonomous robots. Use HMI to confirm the robot's feedback and implement corrective action as a Human/Machine team, rather than to be fully automated. As for robots which send back info from far away (e.g. the drones) there's a time lapse in there which could affect decisions adversely....the physical distance between the controller and the robot needs to be reduced since there are at least 8 time dependent signal "journeys" between sending info, receiving it, deciding a course of action, transmitting it back and when the robot gets the instruction for it to trigger the action...by that time, however small, the circumstances could have changed, even in the case of a drone which has locked on to a specific target.
ScotCan, can you define what you mean by "fully autonomous"? Autonomous robots already exist. Some of them have the option of being controlled remotely, and many can send back data to a remote human, using various forms of communication.
That robot running at 28MPH was very impressive. It was more like a gallop than a run, though. Not quite horse style yet, and it looks like "horse style" would have a better ability to balance. A robot as stable as a horse would certainly have a whole lot of applications. It might even be useful in getting around city traffic jamups. And the military uses would be totally demoralizing to the enemy. Just imagine, if the robots were dressed as soldiers, running in a charge, firing automatic weapons with both "hands". That would make almost everybody drop and run.
The Navy robot probably has a Human/Machine Interface since once the robot has entered the fire zone the decision process would be handled by the human supervisor. Since the robot is working locally it makes sense to have an umbilical cord carrying all the relevant data back to a central control...a fully autonomous robot is still down the pike a bit.
I think it's an understandable bias, Ann. If we're trying to get a machine to replicate human movement -- as with the fire fighting robot -- it makes sense that we work with the solutions we already know and understand, our own movements.
Yes, the crawlers may make more sense when it comes to movement. I still keep thinking there is a bias toward robots with human attributes -- like legs. Replicating human movement may not make the most sense.
Rob, that's a really good point. We featured crawling robots in the Bugs and Worms robot slideshow:
and some of them, as well as other, snakelike robots, do workarounds and learn. I think the problem with the legs versions is that they're more likely to tip over because of a much higher center of gravity.
Festo's BionicKangaroo combines pneumatic and electrical drive technology, plus very precise controls and condition monitoring. Like a real kangaroo, the BionicKangaroo robot harvests the kinetic energy of each takeoff and immediately uses it to power the next jump.
Design News and Digi-Key presents: Creating & Testing Your First RTOS Application Using MQX, a crash course that will look at defining a project, selecting a target processor, blocking code, defining tasks, completing code, and debugging.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.