Dynamic Multi-Objective Ergodic Search
Robotic explorers are essential tools for gathering information about regions that are inaccessible to humans. For applications like planetary exploration or search and rescue, robots use prior knowledge about the area to guide their search. Ergodic search methods find trajectories that effectively balance exploring unknown regions and exploiting prior information.
In many search based problems, the robot must take into account multiple factors such as scientific information gain, risk, and energy, and update its belief about these dynamic objectives as they evolve over time. However, existing ergodic search methods either consider multiple static objectives or consider a single dynamic objective, but not multiple dynamic objectives.
We address this gap in existing methods by presenting an algorithm called Dynamic Multi-Objective Ergodic Search (D-MO-ES) that efficiently plans an ergodic trajectory on multiple changing objectives. Our experiments show that our method requires up to nine times less compute time than a naïve approach with comparable coverage of each objective.
Research Team: Ananya Rao*, Abigail Breitfeld*, Alberto Candela, Benjamin Jensen, David Wettergreen, Howie Choset (*equal contribution)
Left: Scalarized map combining three dynamic objectives (blue, green, and yellow), and the original planned trajectory. Right: Final executed trajectory on the same objectives after re-planning based on change in maps. Each Gaussian peak the robot visits becomes more diffuse, indicating there is less information to be gained in that region once the robot has taken samples there.
First row: Set of objective maps (a-c) and their weighted combination (d). The planned trajectory based on this scalarized map is shown in red. Second row: Updated set of objective maps (e-g) and their new weighted combination and planned trajectory (h). Note that the slope map is a static objective and does not change over time.