Gravity-Free X-Y Acceleration

₹5000-8000 INR

Posted 1 minute ago

•

Ends in 6 days

₹5000-8000 INR

Paid on delivery

Firmware Development: I have an ISM330DHCX IMU on the I2C bus of an STM32H745Zi board. Register access is already running; what I need is the maths. Your task is to fuse the IMU’s gyroscope and accelerometer so the MCU keeps an accurate estimate of its attitude and continuously subtracts the gravity vector. The end result must be two real-time, gravity-free linear acceleration values for the X and Y axes only. The code will sit in a small autonomous-vehicle controller, so it has to stay responsive during sustained vibrations, pothole like movements and slow tilting during slopes and hills—not just brief hand motions. Please keep latency low enough for a 400 Hz control loop and write everything in clean C that compiles inside STM32CubeIDE with HAL drivers. Deliverables • C source and header files implementing the orientation filter and gravity compensation • A simple example task that prints the gravity-free X and Y acceleration over UART every cycle • Short README explaining configuration registers, sampling rate, coordinate frames and any calibration steps Delivery date: 22/11/26 till 10PM Acceptance: when the board is held at any tilt while stationary, reported X-Y acceleration remains within ±0.02 g, and when the platform is driven in a straight 1 m/s² sled test it reports ≈1 m/s².

Project ID: 40816060

About the project

2 proposals

Open for bidding

Remote project

Active 2 mins ago

Place your bid

Bid amount Email address

Benefits of bidding on Freelancer

Set your budget and timeframe

It's free to sign up and bid on jobs

2 freelancers are bidding on average ₹6,750 INR for this job

User Avatar
Flag of GERMANY

Hi, there I recently implemented a Mahony filter on STM32H7 for a drone flight controller fusing ICM-42688 data at 1 kHz, achieving <1 ms latency and stable horizon lock during 30° banked turns and high-vibration motor runs. The main challenge was accelerometer noise aliasing during sustained vibration; I solved this by adding an adaptive gain scheduler driven by accelerometer magnitude variance and a fixed-point implementation to guarantee deterministic cycle time. I am planning to port a fixed-point Mahony AHRS tuned for 400 Hz, integrating ISM330DHCX FIFO watermark interrupts to minimize ISR overhead. I offer to implement a two-stage gravity subtraction: quaternion rotation of the calibrated accelerometer vector followed by a complementary high-pass filter on the linear output to reject low-frequency tilt drift during slopes. Calibration will use a 6-position tumble routine storing scale/misalignment matrices in flash. If I use my previous experience, your project will likely be completed successfully. Hope to discuss this in detail. Through detailed discussion, I think I can find the better solution to finish your project successfully. Thank you

User Avatar
Flag of UNITED KINGDOM

For ±0.02 g at any tilt, the main things are gyro bias and accel offset calibration, and an attitude filter that doesn't trust the accelerometer during vibration or potholes. I'll write a Mahony or complementary-style filter with adaptive accel gain, subtract gravity in the body frame, and output X/Y linear acceleration at 400 Hz in HAL-friendly C. You'll get the source, a UART example task and a README. I've done a lot of STM32 firmware and sensor work, and my portfolio has related industrial boards.

内容来自 Freelancer News,版权归原作者所有。

阅读原文 ↗

把喜欢的内容,留在你的新标签页。

打开我的订阅