Bittle not performing functions properly.

It's disappointing to note that Bittle isn't living up to expectations in terms of functionality. Despite its promise, it seems to be falling short in delivering the intended results efficiently. If you're encountering such issues, it might be worthwhile to seek assistance from experts who specialize in troubleshooting and optimizing Bittle's performance. Consider reaching out to professionals on platforms like Upwork (www.upwork.com/freelancers/~01790af55596a6de17) who have the skills and experience to address these concerns effectively. Don't let Bittle's shortcomings hinder your progress; explore avenues for expert support to ensure smoother operations and better outcomes.

I don't think changed “skillHeader” to “4” would do anything different from the original code.

Yes, every "Y" corresponds to a specific prompt. People are supposed to read through the user manual and find this section: https://bittle.petoi.com/6-calibration#6-1-2-upload-and-run-writeinstinct-ino

I don't quite remember where I put "YYY" in the manual, but I do have the impression that I mentioned it somewhere as a quick start trick.

OK, I’ve got my Bittle up and running and all the skills are working! Very cool. Here’s the compile error I was seeing, very similar to what Gary originally posted, though obviously we have slightly different paths:

In file included from sketch/OpenCat.h:85:0,
                 from /Users/rbolling/Library/Mobile Documents/com~apple~CloudDocs/github/OpenCat/WriteInstinct/WriteInstinct.ino:31:
/Applications/Arduino.app/Contents/Java/hardware/arduino/avr/libraries/Wire/src/Wire.h: In member function 'void Motion::loadDataFromI2cEeprom(unsigned int&)':
/Applications/Arduino.app/Contents/Java/hardware/arduino/avr/libraries/Wire/src/Wire.h:68:13: note: candidate 1: uint8_t TwoWire::requestFrom(int, int)
     uint8_t requestFrom(int, int);
             ^~~~~~~~~~~
/Applications/Arduino.app/Contents/Java/hardware/arduino/avr/libraries/Wire/src/Wire.h:65:13: note: candidate 2: uint8_t TwoWire::requestFrom(uint8_t, uint8_t)
     uint8_t requestFrom(uint8_t, uint8_t);
             ^~~~~~~~~~~
Sketch uses 25758 bytes (79%) of program storage space. Maximum is 32256 bytes.
Global variables use 1372 bytes (66%) of dynamic memory, leaving 676 bytes for local variables. Maximum is 2048 bytes.

Poking through OpenCat.h it seems the trouble was with line 559 where the compiler is confused by the call Wire.requestFrom function. The way I fixed it was to change the parameter “skillHeader” to a straight integer 4 (see below). I also tried another fix that also works by changing the declaration in line 558 to:

int skillHeadder = 4;

 loadDataFromI2cEeprom(unsigned int &eeAddress) {
      Wire.beginTransmission(DEVICE_ADDRESS);
      Wire.write((int)((eeAddress) >> 8));   // MSB
      Wire.write((int)((eeAddress) & 0xFF)); // LSB
      Wire.endTransmission();
      byte skillHeader = 4;
//      Wire.requestFrom(DEVICE_ADDRESS, skillHeader);     //line 559
      Wire.requestFrom(DEVICE_ADDRESS, 4);// <—-changed “skillHeader” to “4”
      period = Wire.read();
      //PTL("read " + String(period) + " frames");
      for (int i = 0; i < 2; i++)
        expectedRollPitch[i] = (int8_t)Wire.read();
      angleDataRatio = Wire.read();

      byte frameSize;
      if (period < -1) {
        skillHeader = 7;
        frameSize = 20;
        Wire.requestFrom(DEVICE_ADDRESS, 3); 
        for (byte i = 0; i < 3; i++)
          loopCycle[i] = Wire.read();
      }
      else
        frameSize = period > 1 ? WALKING_DOF : 16;
      int len = abs(period) * frameSize;
      //delete []dutyAngles;//check here

Once I successfully compiled and loaded I still had to get the skills working. For me, both the manual section 6 on Calibration and Dobromir’s advice to type “YYY” were a bit misleading. Though both are essentially correct, you don’t just type “YYY” you actually must do each of those calibrations. If you type a Y at the servo calibration it zeros out any calibrations you did before. I’ve learned to write down my settings. This saves having to redo the tedious, but necessary servo calibration using the “L” gauge, I just re type in my numbers. Next the very first time you run WriteInstincts you HAVE to answer “Y” to update the instincts otherwise the instincts never get loaded. The manual is a bit misleading here because it implies you only say “Y” if you’ve changed anything. That’s true but you also have to say “Y” if it’s you first time through. Next you really have to calibrate the sensor with the Bittle body flat, all servo legs move out and up out of the way otherwise he’s a bit unstable.

Now own to teaching Bittle new tricks!

This is what my upload of WriteInstinct looks like. However I now have 90% of the functions working. This time I closed Arduino IDE and went to Windows File Manager and double clicked on WriteInstinct.ino. Before I opened WriteInstinct.ino from within Arduino IDE using file_open. I also did this with OpenCat.ino. I answered YYY to both. After recalibrating I still get the above messeges but things seem to work. Any idea what the red messages mean?

Hope this helps Robert.

I am having the same issue. Although my path names are a bit different, the unusual "uint8_t" console output in red but without any specific error and bizarre behavior is the same. Crossing my fingers to solve this. Happens with with both OpenCat and WriteInstinct INOs. Doesn't seem to be what Rongzhong Li is discussing.

The WriteInstinct.ino is only used for saving parameters onto the board. You need to upload OpenCat.ino for proper functionality.

The bluetooth app uses an old token set and doesn't support kwkF, but you can add a customized robot and define all the command keys by yourself.