Extended Kalman Filter for Motion Sensors

For the past three years I had worked in a small team on an quadrotor UAV project. All software I have created was ran using STM32F4 microcontrollers (STM32F405, STM32F439 and STM32F446). I have developed, inter alia, Kalman filter and extended Kalman filter-based algorithms for fusion of gyroscope, accelerometer, magnetometer, barometer, GNSS and rangefinder (sonic and laser) data. Right now I am starting my own business with aim of providing software and system design engineering services. I will be glad to answer any of your questions either here or on Skype (live:juda.piotr_1). Best regards, Piotr

Hello, I am interested in your project. I have a lot of experience with STM32, accelerometers, gyroscopes, GPS. I have a big project for navigation. I have all the necessary development tools.

Professional aerospace engineer with experience in Avionics software development and DSP........................Prefer working at $66/hr rate with $660 as Initial MILESTONE

Hello! I am extremely interested in this project and would love to work on this project, please reply back to discuss further. Looking forward to hear from you soon. Thanks!

Dear sir, I have already implemented Extended Kalman filter filter for GPS/INS integration. It used IMU data - 3 axis accelerometer, gyro and magnetometer with as initial measurements and GPS and barometric pressure as aids for filtering. I have lots of experience developing Kalman filter algorithms with different nature. I can implement aerospace grade kalman filter algorithm similar to which is used in VN200 sensor. As you mentioned UKF, EKF is the standard way of implementing GPS/INS filtering. All most all industrial GPS/INS sensors uses EKF in loosely coupled configuration due to its stability with numerical errors(fixed point arithmatic). The price I bid is for All below features. Aerospace grade Kalman filter with INS sensors + GPS + Barometric pressure. This will have 15 states. Please drop me a message for further discussion. Thank you.

I have had experiene with Kalman Filter (on orbital calculations using GPS data). I can do this but for optimal results I would need access to the sensors you use. That is the reason I set 80 days as the delivery time. Alternatively you can collect the data from all the sensors during a linear accelaration and send it over, it would speed things up. But you would need someone on your end to do the final implementation on your target board. I would do this in C, it would be easy to port but it might require some additional optimizations for your board as this filter might end up consuming quite a bit of resources. I can not predict it before the first prototype. Regards

As we are a team of computer engineers, we already have a solid experience with firmware and embedded software. Besides, we gather knowledge in filters and PDS.