Optimal Vehicle Behavior
Contact person: Lars Nielsen
Hybrid look ahead controlLook-ahead control deals with the problem of driving a vehicle in a fuel optimal way that benefits from information of the road ahead. Since it is nowadays possible to know the topography ahead using GPS and a map data base, such fuel optimal controls are being deployed for on-board vehicle control. In this project the word hybrid has double meaning. Firstly, it aims at hybrid vehicles equipped both with a combustion engine and an electrical motor, each with their respective energy storage (fuel tank and batteries). Secondly, the word hybrid also refers to the methods needed for modeling and optimization. The models contain gear shifts and other mode switches between combustion and electrical thus being hybrid models in the sense of systems science. The models also include non-linear look-up mappings. The optimization methods have to handle this complexity. Three phases are illustrated: in first position lift of accelerator, in second roll to accelerate, and finally in third position accelerate to enter at higher speed.
Developing the new vehicle propulsion laboratory
A new building, L-huset, has been built and it was inaugurated October 5, 2011. Within the building there are three laboratories, namely the laboratory for vehicle propulsion, the laboratory for vehicle informatics, and the laboratory for energy storage and transformation.
Important for Cadics is the four wheel vehicle dynamometer that was funded by Moviii. That laboratory is developed and extended together with Rototest and VTI. It is used for measurements and for simulator experiments. Two important current research projects regard drive cycles and vehicle behavior.
Drive cycles in vehicle optimization
Drive cycles are used for vehicle testing and certification, but also as an engineering tool during development. In the latter application an important question is what test cycle to use to be able to excite and examine a certain component or function under study. In general terms the aim is to establish the correspondence between drive cycle and test object. Two projects that have been performed so far are a study of co-surge in a twin turbo V-engine car, and a study of the split axle hybrid on the photo. A diesel drives the front axle and an electrical motor drives the rear axle.
Vehicle behavior models
The vehicle behavior part of the project combines the propulsion laboratory with the informatics laboratory in order to find efficient descriptions. The picture below shows one of our cars equipped with additional modern sensing equipment on the race track we collaborate with to obtain measurements.
Projects so far has been about powertrain oscillation analysis aiming at tire pressure analysis, and models for at-the-limit maneuvers. The latter models have been used for calculations of time-optimal behavior.
Road geometry and vehicle motion estimation
We are working on a sensor fusion solution for the problem of joint ego-motion and road geometry estimation by making systematic use of measurements from a forward looking radar and camera (infrared or standard), steering wheel angle sensor, wheel speed sensors and inertial sensors. We use dynamical models for the ego vehicle, the road and the leading vehicles. The proposed solutions have been tested and compared to existing algorithms, using measurements from authentic traffic environments on public roads in Sweden.