Controlling the engine load of combine harvesters is a difficult task due to quickly varying field conditions, numerous machine configurations and delayed sensing. The controller presented in this work pro-actively reacts to varying field conditions by using geospatial information derived before the crop is sensed inside the machine. For example, this information may consist of yield or crop height estimates. A model predictive control algorithm is chosen to incorporate varying biomass predictions and to determine a speed planning for the next sample steps. This speed planning is used to transform geospatial data into the time-domain such that it can be consumed by the control algorithm. The dynamic behavior and the power distribution of a combine harvester are modeled offline and adapted online to account for different machine configurations and non-linearities. The advantage of using anticipative knowledge is shown in simulation and field test data.
Model Predictive Control for Combine Harvesters using Geospatial Data
Proceedings
2021-04-18
14 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Deutsch
Electro-Hydraulic Header Control for Combine Harvesters
SAE Technical Papers | 1994
|Electro-hydraulic header control for combine harvesters
Kraftfahrwesen | 1994
|INTELLIGENT HYBRID POWERTRAIN SYSTEMS FOR COMBINE HARVESTERS
Europäisches Patentamt | 2022
|Investigation of a feedrate sensor for combine harvesters
Kraftfahrwesen | 1993
|