Episode 14 Discusio...
 
Notifications
Clear all

Episode 14 Discusion  

Page 3 / 3
  RSS

Duce robot
(@duce-robot)
Reputable Member
Joined: 11 months ago
Posts: 454
2020-03-24 12:10 am  

@robo-pi

i hate to be a bother but if i was loading software on to the nano for sweep lidar for ros where would i put it ? by the way the module is by makerfocus and is $73 pretty cheap on amazon only 14 left if you want one but would it just go to downloads? its a nice unit check it out.


ReplyQuote
Robo Pi
(@robo-pi)
Robotics Engineer Moderator
Joined: 12 months ago
Posts: 1757
2020-03-24 2:24 am  
Posted by: @duce-robot

i hate to be a bother but if i was loading software on to the nano for sweep lidar for ros where would i put it ?

I'm afraid I have no clue what you are talking about. 😊  I'm no expert on Linux systems.  I typically just use sudo apt-get install xxxxxxx and the system just installs it.  I don't recall having to tell the system where to put things specifically.   On rare occasions when I need to do a manual install there are usually instructions with the software that suggests a directory to install it in.   In short, I would need more information about exactly how you are installing this software.   Did it come with a readme file?

Posted by: @duce-robot

by the way the module is by makerfocus and is $73 pretty cheap on amazon only 14 left if you want one but would it just go to downloads? its a nice unit check it out.

I already have a TF-Mini Lidar that I haven't even had time to try out yet. 😊 

I've got my hands full right now trying to organize other projects.  I'm trying to set up pyttsx3 using espeak and MBROLA on the raspberry Pi 4, and that's turning into becoming a major study.   In fact, it's so complicated I'm thinking of making an entire video series on it.

I'm also cleaning up some of my basic system SD cards for both the Raspberry Pi and the Jetson Nano.   So I'm reloading those systems from scratch which is taking a lot of time but will be well worth it in the end.

I recently bought a Raspberry Pi 4 and installed Ubuntu on it.   It's looking to be quite promising.   If it works out well I think I'll buy another one. 😎  Especially when I can just plug in an already configured system on the new Pi.  I really like these little SBC's with SD system cards instead of hard drives.  It makes it so much easier to just duplicate an entire system instantly by just making a copy of the SD card.   So I'm becoming an SBC enthusiast here.

 

DroneBot Workshop Robotics Engineer
James


ReplyQuote
Duce robot
(@duce-robot)
Reputable Member
Joined: 11 months ago
Posts: 454
2020-03-24 3:03 am  

Yes the SBC can be addicting I have several but yes that is what I'm assuming sudo apt get install I'll find out soon its made to use with ros which I have not gotten in to yet but I loaded it but have not used melodic is a little complicated for me as of yet but in time I'll grasp it the espeak sounds interesting make the videos I'll watch them . I want to try for speech eventually.I'm already in too deep might as well go for broke lol 👍😁


ReplyQuote
Robo Pi
(@robo-pi)
Robotics Engineer Moderator
Joined: 12 months ago
Posts: 1757
2020-03-24 7:19 am  

I have absolutely zero knowledge of ROS, and that's not likely to change anytime soon.  I have nothing against ROS.  I simply don't have any immediate use for it, and I have other things that I currently consider to be far more important for me personally.   I'm having difficulty even finding the time for the things I would really like to learn in more detail.

I'm currently studying four areas of study:

  1. IMU Inertial Guidance Systems using PID and quarterions.
  2. Perceptrons and Neural Networks
  3. Semantic A.I.
  4. FPGA and CPLD design.

All of these topics are also causing me to necessarily need to study even more topics that are sub-related.  In particular at the moment, the Semantic A.I. project is taking me down a road of learning the details of Speech Engines, both TTS (Text to Speech) and SRE (Speech Recognition Engines).   I could actually study Semantic A.I. using just plain text.  Have the computer print out its responses, and type in what I want to say to it.  But I'd rather do it verbally.   I'm also discovering that software along those lines is not easy to come by.  Especially if you want something that is fully programmable and not just canned speech.  So I've decided to go all-out with the speech engines.

And then on top of all that I'm hoping to make videos to share everything I'm doing on all four of these subjects.  So I'm learning video production too.  No time for ROS.  😊

Today, I spent the entire day making a brand new system SD card for the Raspberry Pi 4.  Of course, this includes the installation and customization of all the software I need for studying the subjects mentioned above.   But I think this is time well spent.   Getting the core system set up so it's exactly how I want it will make the actual studies go much smoother.

In any case, I wish you the best with ROS.    It's just not a topic that I can speak to.   I have no experience with ROS, and don't expect to obtain any experience anytime soon.  About the only way I'll learn about ROS is if Bill decides to make a video on it.  And even then I'd just watch the video, and and that's about it.

DroneBot Workshop Robotics Engineer
James


ReplyQuote
Duce robot
(@duce-robot)
Reputable Member
Joined: 11 months ago
Posts: 454
2020-03-24 2:02 pm  

I wonder if you could make your own speech engine with an SD card and a mp3 with the vocabulary on the SD card with the robot being able to choose responses from the SD card . but as for ros I'll get it its just going to be slow .but its OK I'm still learning  and the robot is already performing  most of the functions ,and let's face it if the duce is moving around you can't play with all the toy's on it  its more like a work station - robot-disco tech- security system  but yes speech is something I want in it at some capacity even if it is canned speech  I think it will bring another cool element to it if you do some tutorials on speech I'd love to see them.👍  in the mean time I'm going to release a new video premiere of the duce robot and doctor s operation midnight enter control  X  it might break things up a bit and maybe even get a laugh or two.👍😁


ReplyQuote
Robo Pi
(@robo-pi)
Robotics Engineer Moderator
Joined: 12 months ago
Posts: 1757
2020-03-24 11:04 pm  
Posted by: @duce-robot

but yes speech is something I want in it at some capacity even if it is canned speech  I think it will bring another cool element to it if you do some tutorials on speech I'd love to see them.👍

If you just want a talking robot there are far simpler ways to do it than what I am working on.   Also, if you're willing to use Windows as the OS, you can simply use Microsoft Speech Platform which already works very well.   However, that's also dependent upon using C#.   And that's ok as long as you don't mind being tied to those tools.   There are also many other speech platforms available if you are open to using Internet based speech engines.   You can even use chatbots like IBM Watson, Alexa, and others.  These are all very easy to use  if you're just looking for a talking robot that can also recognize speech.

I've chosen to take a much different road for my A.I. project.   Including requiring the following criteria.

  1. Must be free of any Internet dependency
  2. Must be free from any specific OS platform
  3. Must be compatible with Python
  4. Must also be based on phonemes instead of prerecorded words.

So it's meeting all these criteria that make my project somewhat difficult.   I did find speech engines that meet these criteria, but unfortunately they are also quite crude and so will require a lot of refinement on my part.

By the way, I already have a speech system set up using Windows  and C#.   I hope to be making a video on that as well.  So we'll see where all this leads.

DroneBot Workshop Robotics Engineer
James


mbogelund liked
ReplyQuote
Page 3 / 3