# US 12,169,777 B2 · Artificial-intelligence-based waterway information system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,169,777 B2 - Document ID: US12169777B2 - Status: Granted - Kind: B2 - Application number: 18/128,839 - Filed: 2023-03-30 - Published: 2023-07-27 - Granted: 2024-12-17 - Earliest claimed: 2020-04-28 - Assignees: Trabus - Applicants: Trabus - Inventors: Joseph Celano; David Sathiaraj; Eric Ho; Andrew Nolan Smith; Eric Vincent Rohli - Web page: https://patentgrep.com/patent/US12169777B2 ## CPC - G06N 3/08: Learning methods - B63B 79/40: for controlling the operation of vessels, e.g. monitoring their speed, routing or maintenance schedules - G01C 21/203: {specially adapted for water-borne vessels} - G06N 3/04: Architecture, e.g. interconnection topology - G06N 3/045: Combinations of networks - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/09: Supervised learning - G06Q 10/047: Optimisation of routes or paths, e.g. travelling salesman problem - G08G 3/00: Traffic control systems for marine craft (marking of navigational route B63B22/16, B63B51/00) - B63B 2213/00: Navigational aids and use thereof, not otherwise provided for in this class - G06N 20/20: Ensemble learning - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - B63B: SHIPS OR OTHER WATERBORNE VESSELS; EQUIPMENT FOR SHIPPING (arrangements of vessel ventilation, heating, cooling, or air-conditioning B63J2/00; floating substructures as supports of dredgers or soil-shifting machines E02F9/06) - G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S) - G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR - G08G: TRAFFIC CONTROL SYSTEMS (guiding railway traffic, ensuring the safety of railway traffic B61L; arrangement of road signs or traffic signals E01F9/00; radar or analogous systems, sonar systems, lidar systems specially adapted for traffic control G01S13/91, G01S15/88, G01S17/88; {radar or analogous systems, sonar systems, lidar systems specially adapted for anti-collision purposes G01S13/93, G01S15/93, G01S17/93}) ## Abstract Artificial-intelligence-based river information system. In an embodiment, a first training dataset is used to train a travel time prediction model to predict a travel time along the waterway for a given trip. In addition, a second training dataset is used to train a river level prediction model to predict a river level along the waterway for a given time. For each of a plurality of trips, a request is received that specifies the trip and a time of the trip, and, in response to the request, the travel time prediction model is used to predict a travel time for the trip, and the river level prediction model is used to predict a river level of the waterway at one or more points along the trip. Then, a voyage plan is generated based on one or both of the predicted travel time and the predicted river level. ## Claims ### Claim 1 A method comprising using at least one hardware processor to: use a training dataset, comprising positional and non-positional data representing trips with known travel times along a waterway, to train a travel time prediction model to predict a travel time along the waterway for a given trip; for each trip of one or more trips by a maritime vessel along the waterway, receive a request that specifies that trip and a time of that trip, and in response to the request, use the travel time prediction model to predict a travel time for that trip, and display a representation of that trip on a virtual map within a graphical user interface with an indication of the predicted travel time. ### Claim 2 (depends on 1) The method of claim 1, wherein the travel time prediction model comprises a gradient boosting model. ### Claim 3 (depends on 2) The method of claim 2, wherein the gradient boosting model comprises an XGBoost model. ### Claim 4 (depends on 3) The method of claim 3, wherein training the XGBoost model comprises optimizing parameters of the XGBoost model using a random search. ### Claim 5 (depends on 2) The method of claim 2, wherein training the travel time prediction model comprises applying cross-validation using mean absolute percentage error. ### Claim 6 (depends on 1) The method of claim 1, wherein the travel time prediction model comprises a support vector machine. ### Claim 7 (depends on 1) The method of claim 1, further comprising generating the training dataset, wherein generating the training dataset comprises correlating positional data, indexed by first identifiers of vessels, with non-positional data, indexed by second identifiers of vessels. ### Claim 8 (depends on 7) The method of claim 7, wherein the first identifiers are Maritime Mobile Service Identities. ### Claim 9 (depends on 8) The method of claim 8, wherein the second identifiers are U.S. Coast Guard identification numbers. ### Claim 10 (depends on 1) The method of claim 1, wherein the positional data comprise, for each of a plurality of vessels, information representing trips that the vessel completed along a waterway. ### Claim 11 (depends on 10) The method of claim 10, wherein the positional data comprise Automatic Identification System (AIS) data comprising a plurality of AIS messages, and wherein generating the training dataset further comprises creating a plurality of training trips from the positional data and non-positional data by, for each of the plurality of vessels, bundling a subset of the plurality of AIS messages, in which the vessel is identified, together into one or more trips, based on one or more criteria. ### Claim 12 (depends on 11) The method of claim 11, wherein the one or more criteria comprise: each subset must comprise a plurality of AIS messages; an AIS message is not added to a non-empty subset unless either a time interval between the AIS messages and an AIS message in the subset is less than or equal to a first predefined threshold, or the time interval is less than a second predefined threshold, which is greater than the first predefined threshold, and the vessel traveled at least a predefined speed during the time interval; and a training trip ends whenever either a speed of the vessel becomes zero, a direction of the vessel changes between upstream and downstream, or the vessel stops at a lock. ### Claim 13 (depends on 1) The method of claim 1, wherein the non-positional data comprise, for each of the plurality of vessels, one or more attributes of the vessel. ### Claim 14 (depends on 13) The method of claim 13, wherein the one or more attributes of the vessel comprise at least one physical dimension of the vessel. ### Claim 15 (depends on 1) The method of claim 1, wherein the travel time is an estimated time of arrival. ### Claim 16 (depends on 1) The method of claim 1, wherein the waterway is logically divided into a plurality of channels that are logically connected via a directional graph network, and wherein the travel time prediction model comprises a separate model for each of the plurality of channels. ### Claim 17 (depends on 1) The method of claim 1, wherein the training dataset further comprises statistical features of the waterway. ### Claim 18 (depends on 1) The method of claim 1, further comprising, for each of the one or more trips, updating the indication of the predicted travel time, in real time, based on real-time predictions of the travel time for the trip by the travel time prediction model. ### Claim 19 A system comprising: at least one hardware processor; and software that is configured to, when executed by the at least one hardware processor, use a training dataset, comprising positional and non-positional data representing trips with known travel times along a waterway, to train a travel time prediction model to predict a travel time along the waterway for a given trip, and for each trip of one or more trips by a maritime vessel along the waterway, receive a request that specifies that trip and a time of that trip, and in response to the request, use the travel time prediction model to predict a travel time for that trip, and display a representation of that trip on a virtual map within a graphical user interface with an indication of the predicted travel time. ### Claim 20 A non-transitory computer-readable medium having instructions stored therein, wherein the instructions, when executed by a processor, cause the processor to: use a training dataset, comprising positional and non-positional data representing trips with known travel times along a waterway, to train a travel time prediction model to predict a travel time along the waterway for a given trip; and for each trip of one or more trips by a maritime vessel along the waterway, receive a request that specifies that trip and a time of that trip, and in response to the request, use the travel time prediction model to predict a travel time for that trip, and display a representation of that trip on a virtual map within a graphical user interface with an indication of the predicted travel time.