Notifications
Clear all

Sensor use/purpose  

Page 4 / 4

byron
(@byron)
Reputable Member
Joined: 2 years ago
Posts: 450
 
Posted by: @briang

I documented it here:

That is a very interesting video and its shaping up to be a good project to follow.  I did not see any links to documentation though, so nothing to copy for my snippets file 🤨 for my future bot.  Maybe you are only intending to show video's, and if so then please ignore comment on the documentation, and I will just enjoy the vids. 😀 


BrianG liked
ReplyQuote
BrianG
(@briang)
Eminent Member
Joined: 1 month ago
Posts: 47
 

@byron

Sorry, I was more or less referring to the build of the physical robot for this one and the overview of the controller algorithms and how they work together.

Like the previous 2 videos that show how manage dc motors and ultrasonic sensors working up to an object following behavior in the 2nd video, I will be continuing to create videos that attack each part of the problem.  Soon will be a wall following behavior, using odometry to track robot location, goal behavior, obstacle avoidance behavior, and then finally putting them together for the full algorithm.  I have to let my class catch up as the videos are released around the course schedule.

I should add.. Once I get to potential fields for the low level control then the video series is going into mapping / planning and the other high level parts I talked about in this video.

/Brian


ReplyQuote
robotBuilder
(@robotbuilder)
Honorable Member
Joined: 2 years ago
Posts: 654
 

@briang

Thank you for sharing. Now I am seriously thinking about getting back into my own similar project. I share your views if not your expertise.

Personally I think robot navigation and positioning by the environment rather than by encoders alone is the way to go.

Too zippy for your furniture 🙂 I notice there are no bumper bar sensors? Even something as simple as foam covered bumpers should work to protect furniture? Plastic is good vs. metal. I think such sensors are needed as a fail safe should the higher sensors fail or get it wrong.

 

 


BrianG liked
ReplyQuote
BrianG
(@briang)
Eminent Member
Joined: 1 month ago
Posts: 47
 

@robotbuilder

You are absolutely correct, I need to find time to add bumpers!  My head is always so far ahead of reality.. At the very least I will find something like foam to soften the blows 🙂

I look forward to seeing your project evolve! 

/Brian


ReplyQuote
robotBuilder
(@robotbuilder)
Honorable Member
Joined: 2 years ago
Posts: 654
 

@briang

Possibly using a stall condition detected by increase current demand will be sufficient to indicate that the robot base has been stopped by an obstacle? The encoders would stop sending pulses as well. If the motors spin on the spot a third swivel encoded wheel could be used to flag the robot was not moving and thus has hit an obstacle.

 


BrianG liked
ReplyQuote
SuperCharlie
(@supercharlie)
Trusted Member
Joined: 1 month ago
Posts: 90
 

@briang

Good stuff man!


BrianG liked
ReplyQuote
BrianG
(@briang)
Eminent Member
Joined: 1 month ago
Posts: 47
 
Posted by: @robotbuilder

@briang

Possibly using a stall condition detected by increase current demand will be sufficient to indicate that the robot base has been stopped by an obstacle? The encoders would stop sending pulses as well. If the motors spin on the spot a third swivel encoded wheel could be used to flag the robot was not moving and thus has hit an obstacle.

 

Yes!  I like this idea.. Might even see if i can take it further and determine if the expected encoder data is different from the actual readings.  Might actually be able to make the robot "touch sensitive" 

I feel a video coming on.. 🙂

/Brian


huckOhio liked
ReplyQuote
Page 4 / 4