Kalman Filter and Quaternions
hello,
i in process of constructing vertical take-off , landing vehicle utilizes 6-solenoid carbon dioxide attitude control system.
for initial set-up, interested in controlling 3-rotational degrees of freedom. using arduino uno coupled mpu-6050 gyro.
i have read threads in terms of programming mpu-6050 , using kalman filter. there several questions have been left un-answered , kindly request other users knowledge in area can provide me guidance.
1. after activate dmp (digital motion processor) , extract quaternion, there 2 ways go it. can use "unscented kalman filter" directly on quaternion , move on finding "acs reaction vector". other way resolve quaternion pitch, yaw, , roll , apply simple kalman filter separately each angle. best way go it?
2. once calculate reaction vector, need obtain right combination of acs thruster bursts. acs needs balance external forces, remove unwanted rotation, , return vehicle original orientation. general structure of code, in other words, set "if statements" in relation current acceleration, velocity, or angle?
thank time...
i in process of constructing vertical take-off , landing vehicle utilizes 6-solenoid carbon dioxide attitude control system.
for initial set-up, interested in controlling 3-rotational degrees of freedom. using arduino uno coupled mpu-6050 gyro.
i have read threads in terms of programming mpu-6050 , using kalman filter. there several questions have been left un-answered , kindly request other users knowledge in area can provide me guidance.
1. after activate dmp (digital motion processor) , extract quaternion, there 2 ways go it. can use "unscented kalman filter" directly on quaternion , move on finding "acs reaction vector". other way resolve quaternion pitch, yaw, , roll , apply simple kalman filter separately each angle. best way go it?
2. once calculate reaction vector, need obtain right combination of acs thruster bursts. acs needs balance external forces, remove unwanted rotation, , return vehicle original orientation. general structure of code, in other words, set "if statements" in relation current acceleration, velocity, or angle?
thank time...
that sounds challenging project.
given you're tackling it, assume know kalman , role in solution. in general expect feed whole sensor input kalman filter , resolve modes @ same time haven't ever implemented in microcontroller don't know whether there practical issues require different approach. google shows various projects applying kalman filters imu perhaps further reading useful - example, https://github.com/tkjelectronics/example-sketch-for-imu-including-kalman-filter/tree/master/imu6dof/mpu6050.
given want control movements around 3 axes guess pid each degree of freedom way go. worth time research control algorithm used on multirotor uav projects see whether have found better approach.
if aren't committed implementing yourself, think save huge amount of effort adopting 1 of open source uav autopilot projects starting point project.
given you're tackling it, assume know kalman , role in solution. in general expect feed whole sensor input kalman filter , resolve modes @ same time haven't ever implemented in microcontroller don't know whether there practical issues require different approach. google shows various projects applying kalman filters imu perhaps further reading useful - example, https://github.com/tkjelectronics/example-sketch-for-imu-including-kalman-filter/tree/master/imu6dof/mpu6050.
given want control movements around 3 axes guess pid each degree of freedom way go. worth time research control algorithm used on multirotor uav projects see whether have found better approach.
if aren't committed implementing yourself, think save huge amount of effort adopting 1 of open source uav autopilot projects starting point project.
Arduino Forum > Using Arduino > Project Guidance > Kalman Filter and Quaternions
arduino
Comments
Post a Comment