Peanut growth and maturity prediction can help farmers and breeding programs improving crop management. Remote sensing images collected by satellites and drones make possible and accurate crop monitoring. Today, empirical relations between crop biomass and spectral reflectance could be used for prediction of single variables such as aboveground crop biomass, pod weight (PW), or peanut maturity. Robust algorithms such as multioutput regression (MTR) implemented through multioutput random forest (RF) regression algorithms capable of predicting multioutput variables has not been proposed for peanut. We developed experiments to predict multiple peanut variables using the MTR approach. The experiment was conducted in 2021 on an 8.5 hectare irrigated peanut commercial field located near Auburn, Alabama. The field was divided into square grids (0.01 hectare size), and 20 grids of contrasting soil characteristics were selected for data collection. Starting 92 days after planning, peanut biomass samples were collected weekly from 1.5 m row length inside each grid. Assessment of peanut maturity was done manually on 200-pod sample using the hull-scrape method and the peanut profile board. Peanut maturity indices (PMI) were calculated using two equations one considering pods from Brown to Black class and other considering Orange to Black classes. To establish functional relationship between peanut biophysical variables and spectral reflectance changes of the canopy over time, Planet Labs imagery data was used to extract vegetation indices (VI) and reflectance from specific spectral bands. The indices NDVI, GNDVI, NLI, MNLI, SAVI and spectral bands were used as explanatory variables. Training (80% of original data set) and cross-validation (20% of data) of algorithms were developed using toolkits available in the Scikit-learn python library. The metric to analyze the performance of the algorithms was the mean absolute error MAE. The RF algorithm outputted multiple numeric values of PMI upon VIs and spectral bands, supporting our hypothesis that MTR can predict the maturity of peanut at field level. The inputting of imagery data from remotely sensing on the peanut canopy produced predictive error of 0.09 % for PMI using brown to black pods and 0.13 % when predicting PMI using orange to black pods. When predicting PW and biomass PW was the variable with more predictability by the framework (MAE = 1039.19) whereas biomass has the highest accuracy (MAE = 892.62). Our findings demonstrated a promising alternative to predict multiple PMI at field scale using remote sensing which may reduce the subjectivity of determine peanuty maturity. Other promising outcome is that predicting peanut biomass above ground and below ground farmers and researches can have quantitative values of those variables allowing characterize the peanut variability throughout the space and time.  Future research should focus on integration of UAV and SAR data to possible improve this method.