Developer

cortex-m7-400mhz.jpg?width=700

Cross-post from my original G+ plus post, based on a The Register article: ARM just announced a new embedded core, the Cortex M7 with up to 400 MHz clock speed. The CPU is in general faster and has an improved pipeline, the real game changer four our drone applications is however the double precision hardware floating point unit. In particular for global coordinate systems (e.g. using GPS) having double precision available in hardware is quite enabling. Interestingly ARM decided to use 3D Robotics (and the Y6 and Iris graphics) and the PX4 project (which has its origins at ETH Zurich) as use case for how their new micro controller could impact the market.

As a a long-time onboard Linux user we are happy to see more and less expensive Linux single board computers becoming available, but interrupt latency, context switch overhead and I/O will remain challenging topics for all-Linux solutions, and so improvements in the low-cost micro controller domain will remain relevant and interesting.

After all, one of the recent novelties claimed by Apple for their iPhone was the addition of their motion processing co-processor, which turns the current iPhone generations into the same design pattern that we have been using a long-time in research: A low-latency IMU/autopilot coupled with a full-fledged mobile computer, combining the best of both worlds in terms of flexibility and latency.

E-mail me when people leave their comments –

You need to be a member of diydrones to add comments!

Join diydrones

Comments

  • this sounds that the pixhawk boards with M7 can be released in relatively short time.
    Compatibility:
    • Cortex M7 is backward compatible with Cortex M4 instruction set
    • STM32 F7 series is pin-to-pin compatible to the STM32 F4 series
  • I hope that ST will also improve the software environment related to stm32. stm32 are good mcu, but programming for it is not as simple as arduino/atmel.

  • Moderator

    @Lorenz,

    Thanks for sharing , good news from ST now we need to wait the new cpu and understand when will be available.On datasheet  read 200 mhz and only 1 mega of flash, but the coremark are very good . At the moment only 1 version is available but is ,too big and bga ... waiting new device then we can try the actual code on new cpu  :)

    best
    Roberto

  • This could definitely allow us to plan new integrations in the current APM code that builds on top of NuttX.

    At the OSRF we are prototyping a deep-embedded ROS on top of NuttX (still early prototype). Possibly, the M7 could be a good point for thinking about APM-ROS convergence.

  • Thanks for sharing Lorenz. 

  • Developer

    Arducopter runs on top of the PX4 middleware, which would be the place that needs adoption. However, ST's press release for the F7 suggests it should be a minor update: http://www.st.com/web/en/catalog/mmc/FM141/SC1169/SS1858?icmp=ss185...

  • would any porting be required for Arducopter code if the ARM M4 is replaced by ARM M7. or would it simply the change of processor.?

  • Developer

    Bill,

    I'm fully aware of the full/small system layout and there are many, many more examples than Symbian OS. There are also POSIX compliant RTOS which could guarantee timing and still offer a Linux-style API. But they both have their limitations. Please also do not forget that neither solution gives you embedded interfaces, which the interesting higher-speed SoCs generally don't offer (you can't do state of the art computer vision on the BeagleBoneBlack which has CAN, and you don't have CAN on the Samsung Exynos 5, which is fast enough for state of the art computer vision). Which is why researchers use the Exynos 5 + autopilot.

    My point is however not limited to hardware or current generation solutions: None of these specialised solutions offer you the flexibility and development speed a COTS Linux environment offers (and the ability to pick any Linux-supported hardware platform). If you combine Linux + external core (which also abstracts the embedded buses / interfaces, which is a huge chunk of the total code size running on an autopilot), you get full flexibility in terms of software, timing and choice of the Linux platform.

  • Developer
    "after all, one of the recent novelties claimed by Apple for their iPhone was the addition of their motion processing co-processor, which turns the current iPhone generations into the same design pattern that we have been using a long-time in research"
    I don't think the design of using a co-processor is new. It's addition does show it has advantages, but its absence is more an indicator of the BOM.
    As for latency issues and context switching, you should look at Symbian OS and it's nano kernel that is capable of running a real-time mobile phone stack and a full multithreaded OS on one arm core.
    http://BOM.As/
  • Moderator

    Could be interesting evolution ... depend of what will be the first chip maker that integrate M7 in its development process . Could be nice if ST will present a new STM32M7 . I see interesting technology from TI R4 that now going at

    http://www.ti.com/lsds/ti/arm/hercules_arm_cortex_r_safety_microcon...

    best

    Roberto 

This reply was deleted.