Research at 成人大片An AI-based system that enables the real-time dispatching of a fleet of vehicles in complex and dynamic environments
Background
The standard practice for developing a software-based fleet dispatching plan, in a mining operation for example, is to choose one goal to optimize and then create a multistage optimization approach such as linear programming or mixed integer linear programming. This approach requires a substantial amount of time to set up and is computationally intensive. As such, standard fleet dispatching plans are static or rigid in nature and difficult to change due to the significant amount of time required to make such changes. In dynamic environments, like mining, standard fleet dispatching plans can create delays and inefficiencies, resulting in increased costs and cycle times. What is needed is a system that can autonomously control the dispatching of a fleet of vehicles in real-time while adapting to changing environmental and operational conditions.
Technology Overview
The technology uses a machine learning algorithm to generate vehicle command signals based on sensor signals received from the vehicles and from those corresponding to one or more states of the environment. Vehicles are provided with one or more sensors selected from, but not limited to, LiDAR, radar, wheel encoder, magnetometer, accelerometer, inclinometer, gyroscope, inertial measurement unit (IMU), global positioning system (GPS), camera (optical and/or infra-red, etc.), proximity sensor, and the like. Sensors can also be present in the environment in which the vehicles operate, for example at starting points, end points or final destinations of the vehicles, to obtain signals indicative of the environmental conditions and to obtain further information about the location and the action of the vehicles. Vehicles are also equipped with communications hardware and software that allows communication between vehicles and with a base station. Communication may be implemented over various types of networks such as, but not limited to, those based on WiFi, global system for mobile communications (GSM), codedivision multiple access (CDMA), long term evolution (LTE), etc.
The technology determines the next action to be completed by a vehicle by taking into consideration the current actions and states of other vehicles, the current state of the vehicle, and one or more current states of the environment, so that an overall operation of the vehicles is optimized for a specific performance goal (e.g., efficiency, productivity, speed of production or completion, maximum output, etc.). The technology continuously optimizes performance in real-time and determines the appropriate actions for each vehicle in a fleet to achieve a desired goal. The technology is particularly suited to implementation with autonomous vehicles, although it can be used with any fleet of vehicles, including conventional, remote operated, or autonomous.
Benefits
The technology provides a dynamic fleet dispatching system that is more flexible than prior approaches and that automates dispatching to increase productivity and efficiency in real-time. The real-time aspect ensures that the decisions are made based on actual positions and states of the vehicles, rather than relying on potentially suboptimal advance planning based on assumptions. In the example of a mine, this dispatch system can control the actions of vehicles related, but not limited to, material delivery, vehicle queues, vehicle maintenance schedules, grade or classification of material loaded, if the vehicle is loaded or not, the current location of the vehicle in the environment, its next destination, its maintenance condition, and the number of vehicles driving to the same location. This technology addresses the issues of fleet dispatch in dynamic environments with changing environmental and operational conditions.
Applications
The technology is applicable to environments where the real-time, autonomous and dynamic dispatching of fleet vehicles is desired. Examples include, but are not limited to, mining, factories and processing plants, construction sites, military operations, disaster zones, etc. The environments can be terrestrial (surface and/or below surface), aerial, aquatic or marine (surface and/or below surface), and extra-terrestrial.
Opportunity
Queen鈥檚 University is seeking industrial partners interested in developing and commercializing the technology. Opportunities for collaborative research also exist.
Patents
- PCT/CA2024/050754 filed in June 2023
IP Status
Patent application submitted
Seeking
- Development partner
- Commercial partner
- Licensing
Posted
September 1, 2024