2
\$\begingroup\$

I've been reading along for quite some time and now I would like to ask a question of my own.

Samsung has a vacuum robot series, with a variety of models that mostly only differ in the software. PCB's etc are the same, besides the new model having an additional WiFi module and some vacuum motors needing an additional motorcontroller.

I myself have the VR20H9050UW, or also called the VR9000, superseded by the 9250, which is actually also the 9350, that only features an additional WiFi-module and a different firmware version.

Since the robots run on open source software (as per http://opensource.samsung.com/opensource/vr9000h/seq/0 ), it should be possible to access the filesystem and for example add the Wireless tools needed for the WiFi module to function, right? We could maybe even build a database with the different software versions (that are somewhat improved in cleaning algorithms overtime) and update our robots ourselves.

Here comes my question: I opened the robot to find out what connection possibilities there are on the PCB, and I found a 4-pin connector (accessible even without opening the robot, by a rubber clip in the housing: leading me to think this is the connection needed to communicate with the software) and an 8-pin connector. I figured the 4-pin would be an UART connection, it says bluetooth on top of the connector, so I took my HC-06 and hooked up the 5V (pin 1) and GND (pin 4) connection. With a multimeter I learned that pin 2 features a constant ±3.3V signal, and pin 3 hovers around 0.1V, thus matching the criteria for RX and TX. However both my HC-06 bluetooth module and my TTL cable don't provide me any data in the terminal with screen/picocom/minicom. I ran a python-script brute forcing the possible baudrates etc, however I got no response there either. Could it be possible that I need to enter a specific mode, before I could connect. Or, since libusb is used, could it be a plain USB-connector or maybe even another protocol/connection?

As to the 8-pin connector: it features 3.3V VCC at pin 1, GND at pin 8 and al the rest in between is 3.3V as well, except for pin 6, which is a steady 3.0V. Could this possibly be a JTAG connector and how could I verify that?

I provided a photograph of the PCB. If it would be of any help, I can also measure the resistance between pins.

enter image description here

If you need any more information, please let me know.

PS: I am just getting to know the different StackExchanges. It's already on the regular Stack Overflow, but I figured it would better fit here. My apologies!

Edit: I have removed the serial-sticker from the chip, which reveals the PCB houses an NXP LPC2926 FBD144. The PCB connected by the black wires at the top of the photograph, which is responsible for giving the power-up/boot command and switching between the different cleaning programs via three touch-responsive buttons, houses an ATMEGA32A.

enter image description here

\$\endgroup\$
3
  • \$\begingroup\$ I don't have enough reps yet to add 2 links in a post. The link to the second photograph is: i.stack.imgur.com/RBtkc.jpg \$\endgroup\$
    – Vincent Bosch
    Aug 13, 2016 at 13:46
  • \$\begingroup\$ Inserted image. \$\endgroup\$
    – JRE
    Aug 13, 2016 at 14:58
  • \$\begingroup\$ You probably won't see anything happening on the UART unless the firmware is already trying to access that port. \$\endgroup\$
    – user2943160
    Aug 14, 2016 at 15:26

1 Answer 1

Reset to default
1
\$\begingroup\$

You should look at other boards in this robot. In front part there is Robot vacuum cleaner (RVC) module from CLABSYS http://www.clabsys.com/html/product_module01.html with:

and sensor board with:

  • Unknown chip (SAMSUNG HART-m310)
  • Front looking camera (it looks up but have fancy mirror to see 180° front view)
  • 5A Brushed DC Motor Driver (TI DRV8840)
  • connector labeled JTAG

I got all Samsung provided source for open source products in VR20H9050UW and there is source for UBOOT, Linux KERNEL 2.6.32, ROOTFS, Linux utilities (alsa, libusb, ncurses, squashfs and zlib), toolchain for building everything and Nexell USB Downloader Driver. So using this USB connection on RVC module there should be a way to upload new firmware. It could be that using same USB connection someone can access already uploaded firmware and extract file system content from it. But we need documentation from CLABSYS (or Nexell) how to do it.

\$\endgroup\$
7
  • \$\begingroup\$ That does sound like a more likely host for the open source components. Probably best to look at that board for a console UART. The USB theory is possible but may not be the case - one thing would be to examine the U-Boot source and configuration and see if it has USB support. Simplest modifications might be to figure out how the ATmega keypad board talks to the rest of the system, snoop that traffic and maybe develop a substitute that can use those functions in different ways. \$\endgroup\$
    – Chris Stratton
    Aug 14, 2016 at 15:27
  • \$\begingroup\$ Thank you for your information! I will have a look at the boards in the front part of the robot. I didn't figure there would be anything else besides some sensors and the camera module of course, especially since the other board is labeled as the main bord, thus leading me to think it would house all the logic. Would you mind sharing the source code provided by Samsung? Or does that violate the terms you had to accept to get it? \$\endgroup\$
    – Vincent Bosch
    Aug 15, 2016 at 15:32
  • \$\begingroup\$ @VincentBosch the right way to get source code is to request it directly from Samsung by writing this request to oss.req[email protected] but I can share download link that I received: RVC_VR9000H.tar.bz2 \$\endgroup\$
    – devegied
    Aug 15, 2016 at 17:06
  • \$\begingroup\$ Thanks for sharing the link. I will see what I can extract from it. In the mean time, I opened up the front of the robot, revealing the sensorboard with JTAG connection and the Clabsys board, with a clearly marked JTAG connection as well. I tried finding some UART port, but I didn't manage to find and connect yet, unfortunatelly. It appears there are indeed USB connections on the Clabsys board. One labeled USB-a and one labeled USB-b, but I think the b-port is actually used via the vias. Haven't identified the 6-pin port yet. Maybe a connection for the WiFi-card found on the 9250/9350 model? \$\endgroup\$
    – Vincent Bosch
    Aug 16, 2016 at 12:16
  • \$\begingroup\$ Here are some pictures I took from the Clabsys-board. On the top, the JTAG connection is clearly marked. Haven't connected to it yet, though. Clabsys-board \$\endgroup\$
    – Vincent Bosch
    Aug 16, 2016 at 12:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.