Collision Avoidance for Electric Mining Shovels via Virtual Construct
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for collision avoidance in operator-controlled machinery, such as electric mining shovels, are inadequate in preventing collisions due to limitations in human operator capabilities and the lack of advanced automated control technologies, leading to potential damage and inefficiencies.
Innovation Solution
A method and system that utilize Model Predictive Control to predict future dynamics of machinery in the environment, generate virtual constructs of the physical space, and adjust operator input commands to avoid collisions by altering trajectory directions or slowing down the machinery, thereby reducing the probability of collisions with other objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automated control systems are used for collision avoidance, then collision prevention capability is improved, but device complexity and cost increase
Solution Approach 1:
The patent introduces a virtual construct (digital twin) as an intermediary between the physical machinery and the control system. This virtual model simulates the physical environment and machinery behavior, allowing collision prediction and avoidance without requiring complex real-time automated control systems. The virtual construct mediates the interaction between operator inputs and physical outcomes, reducing the complexity of direct automated control while improving collision prevention capability.
2Device complexity
If human operators manually avoid collisions, then device complexity is kept low, but collision avoidance effectiveness deteriorates due to human limitations
Solution Approach 1:
The patent creates a virtual copy (construct) of the physical machinery and environment. This copy mirrors the physical system's behavior and allows prediction of collision outcomes without interfering with human operator control. The virtual construct provides collision avoidance assistance while maintaining simple device complexity, as it processes information in the virtual domain rather than requiring complex physical control system modifications.
3Reliability
If operator control is restricted to prevent collisions, then collision frequency is reduced, but productivity and ease of operation deteriorate
Solution Approach 1:
The system performs preliminary collision prediction in the virtual construct before actual collisions occur in the physical system. By predicting potential collisions in advance using the virtual model, the system can alert operators or suggest alternative actions without restricting their control. This preliminary action in the virtual domain prevents collisions while maintaining operator freedom and productivity in the physical domain.
Data Source
AI summary
Systems and methods for reducing the probability of a collision between a first object, whose trajectory is substantially controlled by an operator input command, and a second object. In one embodiment such a method includes receiving an operator input command indicative of first control data; generating model data indicative of a virtual construct of the physical environment proximal to the first object; processing the first control data to predict future dynamics of the first object in the environment; determining whether, on the basis of the predicted dynamics and the model data, the first object is predicted to collide with a second object in the environment; defining second control data for which the first object is not predicted to collide with the second object in the environment; and providing the second control data to a controller coupled to the first object such that the first object is controlled in accordance with the second control data for substantially avoiding the collision of the first object with the second object.