About Telega v1.0 with full support for Cyphalv1 publish date

General
,
1

Hi,
I am building a Nuttx application that requires communication with Myxa B2 ESC via CAN. I am trying to use the new Cyphal v1 for this. For starting, I have been able to exchange some uavcan.node.Heartbeat.1.0 messages between the Nuttx Cyphal node and a Yakut node on different devices. Next, I want to interface and control Myxa ESC. I have updated the ESC firmware to Telega v0.4 hoping that it somehow supports Cyphalv1 but it seems that v0.4 still uses the UAVCAN v0 protocol.
When taking a look at Yakut readme, there seems to be some ESCs (the example with 4 nodes 122-125) that do perhaps experimentally support Cyphalv1 (maybe using soon to be released Telega v1 ? :smiley: ) .

I am wondering if you have any estimation for the release date? Or is there any pre-release bin file that I can play with first? I am in middle of this task so I don’t really want to put it aside and go do sth else unless the release date is too far in the future then I probably need to replan my tasks :slight_smile:

Thank you!

2

Previously I upgraded just the application file , not the bootloader. Now I use the st-link to flash the compound bin file to Myxa. But it now refuses to boot. I think that I am missing the signature as stated here Issues When Upgrading to Telega v0.3 - #7 by mice5 . My UUID at 0x1FFF7A10 is 39 00 2D 00 06 50 30 50 30 34 39 20 C0 F0 00 C0

08:57:50 : Time elapsed during the read operation is: 00:00:00.031
08:58:44 : Memory Programming ...
08:58:44 : Opening and parsing file: com.zubax.telega-1-0.4.8783aada.compound.bin
08:58:44 :   File          : com.zubax.telega-1-0.4.8783aada.compound.bin
08:58:44 :   Size          : 320.33 KB 
08:58:44 :   Address       : 0x08000000 
08:58:44 : Erasing memory corresponding to segment 0:
08:58:44 : Erasing internal memory sectors [0 6]
08:58:47 : Download in Progress:
08:58:54 : File download complete
08:58:54 : Time elapsed during download operation: 00:00:09.715
3

Yes. That is precisely it. There are three reasons why it is not yet published:

  • The new, more robust EKF observer is not yet finished.
  • The documentation is not yet finished.
  • Configuration parameter management requires rework.

The development is taking time because the new motor control core is AUTOSAR C++14 compliant, which adds overheads, and is covered by an extensive test suite both in SITL, HITL, and using automatic on-hardware integration tests.

The release is scheduled for Q3. Meanwhile, we can ship you a preview binary if you are feeling adventurous and is willing to accept the lack of decent docs.

Regarding the signature, I have already replied in PM.

1 Like