Controller design for the Kongsberg Protector remote weapon station
MetadataShow full item record
This thesis is about controller design for the Kongsberg Protector remote weapon station. The Kongsberg Protector RWS is a vehicle-mounted weapon station designed for carrying machine guns and automatic grenade launchers. The weapon station is designed as a four link kinematic chain and is modelled with a set of nonlinear differential equations which are highly coupled. At present, linear PD controllers are used to control the RWS. Theoretically, these controllers are suboptimal for controlling the RWS because the mathematical model of the RWS is nonlinear. Therefore, one of the main objectives of this thesis is to investigate how model-based manipulator controllers can be used on the RWS. As in all mechanical systems, friction is present on the RWS. Stiction is observed on velocity reversals when the joints track low velocities. This leads to reduced tracking performance. Therefore, one objective of this thesis is to suggest designs for friction compensation. The report concludes that the use of model-based manipulator controllers is possible for the RWS in theory under the assumption that the joints of the RWS are not flexible. The motion of the vehicle will reduce the performance of these controllers because the desired trajectory is generated from measurements of the base orientation and because the motion of the vehicle will generate torques on the RWS' joints. These problems can be solved by feed forward of the velocity and acceleration of the vehicle. This requires measurement or estimation of the velocity and acceleration of the vehicle and is dependent on high quality sensors. It is shown that the flexible gears of the RWS causes problems for the feed forward since the gyroscopes are mounted on the load side of the main azimuth gear.The report presents three friction compensation schemes that will identify and compensate for the friction in the joints and motors of the RWS. These methods have been tested by simulation.