US 12,392,624 B2
Stochastic route planning for electric vehicles
Chinya V. Ravishankar, Irvine, CA (US); and Payas Rajan, Riverside, CA (US)
Assigned to The Regents of the University of California, Oakland, CA (US)
Filed by The Regents of the University of California, Oakland, CA (US)
Filed on Jul. 11, 2023, as Appl. No. 18/220,723.
Claims priority of provisional application 63/359,956, filed on Jul. 11, 2022.
Prior Publication US 2024/0019260 A1, Jan. 18, 2024
Int. Cl. G01C 21/34 (2006.01)
CPC G01C 21/3469 (2013.01) [G01C 21/3446 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A method for stochastic electrical vehicle route planning, the method comprising:
storing at a server that is communicatively coupled with one or more vehicles via a wireless communication network, a vehicle road network as a graph comprising vertices and edges between vertices, where the vertices represent locations in the vehicle road network and the edges represent road segments between the locations, where some vertices represent charging stations associated with charging functions representing tradeoffs between vehicle charging times and charge added;
storing at the server for each edge of the graph a first edge weight representing travel time along the edge, and storing for each edge of the graph a second edge weight representing energy consumption for travel along the edge, where the first edge weight is represented as a probability distribution, and the second edge weight is represented as either a probability distribution or a deterministic value representing an expectation of a distribution;
receiving from a vehicle over a communication network, a vehicle origin and a vehicle destination;
performing a Dijkstra search on the graph from the vehicle origin to the vehicle destination, wherein the Dijkstra search computes for each vertex along a path in the network from the vehicle origin to the vehicle destination a convolution of edge weights along the path, wherein the Dijkstra search comprises computing and propagating along the path, labels representing the tradeoffs between charging times and total travel-time, wherein the probability distributions are discretized distributions and wherein the labels computed during the Dijkstra search are propagated per domain interval of the discretized distributions;
transmitting over the wireless communications network to the vehicle a feasible path produced from the Dijkstra search, wherein the feasible path is above a user-specified non-stranding probability, and wherein the feasible path is configured to guide a human driver or an autonomous driving system.