flâneur

The Multiplicative Extended Kalman Filter

matthewhampsey.github.io · 2,837 words · saved by 1 readers

The MEKF is an important modification of the Kalman Filter that makes it applicable to orientation estimation. Unfortunately, when trying to research the topic for multirotor state estimation, I wasn’t able to find a simple (or recent!) explanation. This is my attempt to provide that simple summary. You can see a short clip of it in action on a Beagle Bone Black with a MPU-9250 IMU here: Let’s imagine that you’re trying to write a multirotor control system to stabilise the craft in midair. Or maybe you’re writing software for your VR system of choice and need to estimate the pose of a handheld controller. Or perhaps you’re writing software that will switch a phone’s display between landscape and portrait mode depending on the detected orientation of the phone. In all of these cases, the device will have at least one onboard IMU (with at least a gyroscope and accelerometer and optionally, a magnetometer), and you’ll have to use the measurements from the IMU to derive an accurate estimat

--> The Multiplicative Extended Kalman Filter The Multiplicative Extended Kalman Filter 18 Jul 2020 The MEKF is an important modification of the Kalman Filter that makes it applicable to orientation estimation. Unfortunately, when trying to research the topic for multirotor state estimation, I wasn’t able to find a simple (or recent!) explanation. This is my attempt to provide that simple summary. You can see a short clip of it in action on a Beagle Bone Black with a MPU-9250 IMU here: Setting the scene Let’s imagine that you’re trying to write a multirotor control system to stabilise the craf

saved by

related reading