Patents & Patent Applications
Published European patent applications from optimization, machine learning and mobility services work at BMW Group.
During my work at BMW Group, I contributed to six published European patent applications as a named inventor. They address machine-learning-based ride-request assignment, neural-network-assisted routing, reinforcement learning, and optimization of mobility-on-demand fleets.
Method and system for operating a fleet of vehicles
Density-based fleet optimization to reduce imbalances between vehicle supply and passenger demand across an operating area.
Regional batching technique with reinforcement learning based decision controller for shared autonomous mobility fleet
Adaptive grouping of mobility requests using origin–destination information and reinforcement-learning-based batching decisions for fleet assignment.
Computer-implemented method, computer program and system for assigning a plurality of ride requests to a plurality of vehicles
Two-stage dynamic ride-request assignment combining fast initial heuristics with iterative metaheuristic improvement.
Method and system for routing a plurality of vehicles using a neural network
Neural-network-assisted vehicle routing based on route representations for assigning customer requests to fleet vehicles.
Method and system for routing a plurality of vehicles
Large-neighborhood-search vehicle routing using a machine-learning-trained insertion operator.
Neural-network-based dynamic matching
Learning-based methods for dynamic matching in ride-hailing services, including the selection of metaheuristic algorithm parameters.
These entries are published patent applications, not a list of granted patents. Publication numbers and descriptions link to the corresponding public records; legal status may change over time.