Low Battery Danger

[Humanoido]

Several times when the battery was low, and then indicated low by the flashing of the blue light, RN would go erratic from not enough voltage, and shoot out an arm or a leg, going into some unpredictable unexpected position. I think this is a little dangerous. I'm going to use the battery low circuit and program in a low voltage detection, but not too low, to prevent this sort of thing from happening. I mean we don't want any injuries over this! We want our Robonova to be safe as possible. - Humanoido

[Ric]

This is a natural response a bit like the spasms that occur when somethings about to die.
I think Hitec built this into RN1 as a last ditch attempt to whack an opponent in battle when energy's running out and the chips are down. After all the RN1 is essentially Robo-One competition format.

Ric

[zapod]

I've not experience this with my RN1 - he basically becomes unresponsive and eventually his legs give way as the servos lose power.

However, years ago I had a RC 4-wheel drive buggy with an uprated motor and electronic speed controller. Problem was, as the battery lost it's charge (after 5mins), the electronics would become erratic, sending the buggy in crazy directions at full speed!

[Humanoido]

This is a natural response a bit like the spasms that occur when somethings about to die.

That's an interesting view of how it works! I never expected anyone to say this, but it makes sense. It's starting to sound more human every day. Maybe as we add more complexity to our robots, we'll begin to see natural and unpredictable behaviors.

I've not experience this with my RN1 - he basically becomes unresponsive and eventually his legs give way as the servos lose power.

The legs do give way but it happens last and only after some unexpected spasmodic response. Is there a ghost in the machine or what? Apparently with low power, the code in the controller, and/or the controller itself is randomly getting mixed up and the servos are responding. I wonder why some RNs would respond this way and others not. Has anyone had these experiences??? Maybe there are some differences in the low battery detect software and circuit? If the detection of a higher voltage battery low state happens, maybe the battery discharge is less and RN won't see the spasms.

I'm seriously considering adding another battery pack inside RN. I don't know about anyone else, but an hour of RN operating time seems like only 5 minutes to me. Could we get 2 hours of operating time with paralleled battery packs? But then I'm not sure how they could be charged. Humanoido