LSTM-enabled Level Curve Tracking in Scalar Fields Using Multiple Mobile Robots

Abstract

Autonomous mobile sensor networks are ideal candidates for exploring large-scaleunknown fields with tasks ranging from source seeking, level curve tracking, mapping an unknown field, and many more. In this work, we investigate the problem of level curve tracking in unknown scalar fields using a limited number of mobile sensors. The level curve tracking problem has been studied in many applications such as monitoring the propagation of fire boundaries and the algae blooms. We design and implement a long short term memory (LSTM) enabled control strategy for a mobile sensor network to detect and track the desired level curve. We develop on top of existing research which uses cooperative Kalman Filter as part of its motion control strategy. This existing method is theoretically proven to converge. The LSTM enabled strategy has some benefits such as it can be trained offline on a collection of level curves in known fields prior to deployment, where the trained model will enable the mobile sensor network to track level curves in unknown fields for various applications. So we can train using larger resources to get a more accurate model, while we can utilize a limited number of resources when the mobile sensor network is deployed in the production. We design and implement an LSTM-enhanced cooperative Kalman Filter that utilizes the sensor measurements and a sequence of past fields and gradients to estimates the current field value and gradient. We also design an LSTM model to estimate the Hessian of the field. We utilize these estimates of the field characteristics with motion controllers to track the desired level curve in an unknown field with the center of the sensor network. Simulation results show that this LSTM enabled control strategy successfully tracks the level curve using a mobile multi-robot sensor network.