Virtual EP: a simulator of multi-agents-based virtual plant growth in response to environmental heterogeneity

This paper presents VirtualEP, a novel simulator for eggplant growth, integrating Agent-Based Modelling technology and existing knowledge of plant physiology. VirtualEP simulates the growth and development of eggplant as an evolution of a dynamic branching network whose nodes are represented by Autonomous Virtual Organs (AVOs). The AVO possesses inbuilt data structure, states and functional rules so that it can autonomously perform physiological procedures (eg photosynthesis, nutrient uptake, storage, mobilization and respiration, etc) to respond to environmental heterogeneity. A discrete implementation of pressure-flow paradigm is incorporated to simulate carbon, water and nitrogen transport and allocation among AVOs. Simulation results demonstrate that VirtualEP can effectively deal with global nutrients allocation, growth in response to variation of air temperature, solar radiation as well as water and nitrogen stress. Moreover, VirtualEP can also provide vivid 3D visualization of these features.