How pollinators, such as bees, exploit plant resources is a fundamental question in biology, with deep ecological, economical and societal consequences. When foraging on flowers, pollinators transfer pollen and mediate the reproduction of plants on which most animals (including us humans) rely on. Understanding the spatial foraging strategies and interactions of pollinators across the landscape is thus a critical scientific challenge to discover their influence on plant mating patterns and pollination efficiency. BEE-MOVE uses an interdisciplinary approach to mechanistically link pollinator movements to pollination efficiency at field scales, thereby crossing boundaries between research on pollinator behaviour and plant ecology. We focus on two key pollinators worldwide: the buff-tailed bumblebee and the Western honey bee. Our first objective is to develop a new radar system to record and analyse the individual 3D movements of hundreds of bees foraging simultaneously. Our second objective is to use arrays of communicating radars and robotic plants to study how bees search and exploit food resources in field setups of several square kilometres, by manipulating key environmental factors such as the density of bees, the 3D distribution of plants, and the nutritional content of nectars and pollens. From these observations, our third obkective is to build computational agent-based models to investigate the influence of bee spatial strategies on pollination efficiency. The dialogue between observations and simulations creates a positive feedback towards a robust, multi-level understanding of plant-pollinator interactions at the scale of landscapes. In addition to exploring entirely new grounds in pollination ecology, our results could be used to design practical interventions for conservation, sustainable agriculture and green development in the worrying context of pollinator declines.