Publications listModels of dynamic systems are required for many different reasons. In this project models are of course needed for being able to design observers (Kalman filters). However, to test the observers an inherent part of the design problem is control system design and simulation of the system. These problems too require models of the system under consideration. Depending the application different levels of model complexity are possible, but often it is difficult to know in advance which features are relevant in the model and which are not. It is common to use quite simple models, preferably linear, for observer and control system design while for simulation it is often desirable to use more a complex model.
I this report models of different complexity are derived for the Autonomous Guided Vehicle (AGV) at the Department of Automation. Initially a relatively accurate model is derived and subsequently different simplifications of the model is proposed. The report has been organized in three sections, where the first section by far is the largest: modeling the dynamics, the kinematic transformation and implementation of a simulation model. The models are derived symbolically. The physical vehicle constants can be found in appendix A.
Naturally a compound model describing the vehicle dynamics and kinematics is not enough to design good observers and control systems. To do this it is also necessary to model the all the sensors on the system. However, since the sensors in this case are quite complicated devices it has been found convenient to devote an entire report for this purpose.
For further information, please contact, Finn Kuno
Christensen, IMM, Bldg. 321, DTU 
Last modified Feb 4, 1998
Phone: (+45) 4588 1433. Fax: (+45) 4588
2673, E-mail: fkc@imm.dtu.dk