Cloud-based model deployment and control system (CMDCS) for providing automated driving services
The invention presents a cloud-based model deployment and control system (CMDCS) for providing automated driving services. The CMDCS comprises a cloud-based platform, an onboard unit (OBU), and a Vehicle-to-System component. The cloud-based platform comprises a localization-enhancement subsystem and a cloud computing module. The CMDCS is configured to collect detectable data and undetectable data from vehicles, road, and cloud. Then, the CMDCS deploys a set of end-to-end AI models and methods for automated driving services, comprising sensing, prediction, planning, and control services. The AI models and methods are trained and optimized to process collected data for providing operating parameters for vehicles. Then, the vehicles can be effectively and efficiently controlled and operated by the CMDCS. In addition, the CMDCS is configured to generate and provide detailed time-sensitive vehicle specific control instructions.
1 . A Cloud-based Model Deployment and Control System (CMDCS) comprising:
a) a vehicle comprising an onboard unit (OBU), which comprises a vehicle control module configured to execute control instructions for driving tasks;
b) a cloud-based platform configured to provide information and computing services to support a connected and automated vehicle highway (CAVH) system, and generates and provides detailed time-sensitive vehicle specific control instructions; and
c) a Vehicle-to-System component configured to connect the CAVH vehicle and the CAVH cloud-based platform, and send data and control instructions to the vehicle to support CAVH driving;
wherein said CMDCS provides and performs cloud-based analytic and optimization methods; and
wherein said CMDCS provides vehicle control instructions comprising a longitudinal acceleration rate, a lateral acceleration rate, and/or a vehicle orientation.
2 . The system of claim 1 , wherein the OBU comprises a data collection module configured to collect data from external vehicle sensors and internal vehicle sensors, and to monitor vehicle status and driver status; said OBU collects data comprising: vehicle engine status; vehicle speed; goods status; surrounding objects detected by vehicles; and/or driver conditions to monitor vehicle status and driver status.
3 . The system of claim 1 , wherein said vehicle control module of OBU is configured to execute control instructions for driving tasks comprising car following and/or lane changing; assume control of a vehicle when the automated driving system fails; and/or assume control of a vehicle when the vehicle condition and/or traffic condition prevents the automated driving system from driving said vehicle.
4 . The system of claim 1 , wherein said cloud-based platform comprises:
a) a localization-enhancement subsystem configured to fuse data in the cloud to improve the accuracy of self-positioning; and/or
b) a Cloud Data Computing/Integration/Management component configured to process, integrate, and manage CAVH data in the cloud at the network, corridor, link, node, and vehicle levels.
5 . The system of claim 1 , wherein said cloud-based platform comprises a sensing as a service sub-system, a prediction as a service sub-system, a planning as a service sub-system, a control as a service sub-system, a connectivity as a service subsystem, a privacy as a service sub-system, a security as a service sub-system, and/or a storage as a service sub-system.
6 . The system of claim 5 , wherein event data collection and distribution methods are supported by the Sensing as a Service sub-system, wherein the event data comprise one or more of traffic incident, network congestion pattern, weather event, lane-blocking, special event, breakdown in traffic flow, and/or work zone.
7 . The system of claim 5 , wherein the Prediction as a Service sub-system comprises cloud computing methods for CAVH system prediction and/or analysis, which comprise methods for vehicle activity prediction and/or analysis, which is configured to analyze mechanical status, predicting system failure, and/or predicting optimal speed.
8 . The system of claim 5 , wherein the Planning as a Service sub-system is configured to provide cloud-computing based methods for CAVH operational planning, comprising vehicle operational planning methods, which comprise receiving and/or providing data characterizing departure time, origin, destination, route, vehicle dispatching, vehicle security, and/or emergency planning.
9 . The system of claim 5 , wherein the Control as a Service sub-system comprises cloud-computing methods for CAVH vehicular control, comprising:
a) the methods for vehicle operational control comprise receiving and/or providing data characterizing vehicle speed, direction, map coordinate, and/or spacing relative to other vehicles;
b) the methods for corridor activity control comprise managing automated driving, merging with non-CAV vehicles, and/or interacting with a vehicular ad hoc network (VANET) service;
c) the methods for global activity control comprise receiving and/or providing data characterizing routes, detour, parking, traffic load, congestion; and/or managing first and/or last mile control of vehicles; and/or
d) the methods for system operational control comprise communicating between one or more of an OBU, roadside unit (RSU), Traffic Control Unit (TCU), and/or Traffic Control Center (TCC); optimizing a network; and/or responding to a security and/or emergency event.
10 . The system of claim 5 , wherein said cloud-based platform comprises a Control subsystem, which provides control instructions using a remote component, which is configured to control the CAVH vehicle in response to emergency, theft, physical attack, and/or cyberattack.
11 . The system of claim 1 , which is configured to provide:
a) vehicle data collected by the vehicle OBU from external vehicle sensors and internal vehicle sensors and event data collected by the cloud-based platform;
b) data computation, data integration, and/or data management at various levels, including Network level, Corridor level, Link level, Node level, and Vehicle level;
c) computing and management, and/or analytics and optimization component;
d) onsite and remote control; and/or
e) planning, prediction, control, remote control, and/or optimization to support automated driving.
12 . The system of claim 11 , wherein said data management is configured to manage crowd-sourced sensor data; implement a method comprising managing data from onboard sensors of a vehicle, data from surrounding vehicles, data from road-side sensors, and sensor data shared from other nearby CAVH vehicles; fuse data in the cloud to improve the accuracy of self-positioning, wherein said data comprises one or more of crowd-sourced sensor data, self-positioning data, and/or map data.
13 . The system of claim 11 , wherein said data integration is configured to perform vehicle-level data aggregation and integration, comprising tasks to:
a) analyze and estimate vehicle state, comprising analysis of steering, thrust, and brake data;
b) assist a control processor to execute commands, reduce errors, and disturbances; and/or
c) exchange vehicle state data with CAVH vehicles; and to communicate with non-CAVH vehicles.
14 . The system of claim 11 , wherein said optimization component is configured to comprise:
a) a System Analytics/Optimization component configured to perform System Analytics/Optimization Methods;
b) a CAVH operational optimization component configured to optimize system level tasks; and/or
c) an operational optimization component to analyze vehicle traces using deep learning.
15 . The system of claim 14 , wherein said optimization component is configured to implement:
a) a control signal/logic optimization method comprising providing control signals and/or a logic optimization algorithm to optimize vehicle specific control instructions; and/or
b) a vehicle dispatching and routing management optimization method comprising one or more of minimizing total travel time, minimizing waiting time and deadheading time, and/or increasing system reliability and safety.
16 . A Cloud-based Model Deployment and Control System (CMDCS) comprising:
a) a cloud-based platform configured to provide information and computing services to support a connected and automated vehicle highway (CAVH) system, and generates and provides detailed time-sensitive vehicle specific control instructions; and
b) a Vehicle-to-System component configured to connect the CAVH vehicle and the CAVH cloud-based platform, and send data and control instructions to the vehicle to support CAVH driving;
wherein said CMDCS provides and performs cloud-based analytic and optimization methods;
wherein said CMDCS provides vehicle control instructions comprising a longitudinal acceleration rate, a lateral acceleration rate, and/or a vehicle orientation.
17 . The system of claim 16 , wherein said cloud-based platform comprises a sensing as a service sub-system, a prediction as a service sub-system, a planning as a service sub-system, a control as a service sub-system, a connectivity as a service subsystem, a privacy as a service sub-system, a security as a service sub-system, and/or a storage as a service sub-system.
18 . The system of claim 16 , wherein event data collection and distribution methods are supported by the Sensing as a Service sub-system, wherein the event data comprise one or more of traffic incident, network congestion pattern, weather event, lane-blocking, special event, breakdown in traffic flow, and/or work zone.
19 . The system of claim 16 , wherein said data management is configured to manage crowd-sourced sensor data; implement a method comprising managing data from onboard sensors of a vehicle, data from surrounding vehicles, data from road-side sensors, and sensor data shared from other nearby CAVH vehicles; fuse data in the cloud to improve the accuracy of self-positioning, wherein said data comprises one or more of crowd-sourced sensor data, self-positioning data, and/or map data.
20 . The system of claim 16 , wherein the Control as a Service sub-system comprises cloud-computing methods for CAVH vehicular control, comprising:
a) the methods for vehicle operational control comprise receiving and/or providing data characterizing vehicle speed, direction, map coordinate, and/or spacing relative to other vehicles;
b) the methods for corridor activity control comprise managing automated driving, merging with non-CAV vehicles, and/or interacting with a vehicular ad hoc network (VANET) service;
c) the methods for global activity control comprise receiving and/or providing data characterizing routes, detour, parking, traffic load, congestion; and/or managing first and/or last mile control of vehicles; and/or
d) the methods for system operational control comprise communicating between one or more of an OBU, RSU, TCU, and/or TCC; optimizing a network; and/or responding to a security and/or emergency event.