Escalating Collision Warnings via Velocity Obstacle Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing collision avoidance systems fail to determine the most optimal maneuver in real-time for multiple obstacles and cannot effectively mitigate collisions once they become unavoidable, lacking the ability to provide escalating warnings and feasible options for drivers.
Innovation Solution
A method and system that utilize sensors and an on-board computer to generate linear and non-linear velocity obstacles, providing drivers with escalating severity warnings and determining optimal collision avoidance or mitigating maneuvers by analyzing the velocity vectors of both the driven vehicle and detected obstacles, activating safety accessories like airbags when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If computer generated avoidance maneuvers are introduced for automated vehicles, then the rate of safe collision avoidance maneuvers increases, but the system complexity and computational requirements increase significantly
Solution Approach 1:
The patent segments the collision avoidance problem into distinct phases: detection phase (sensing obstacles), prediction phase (generating velocity obstacles), decision phase (selecting maneuvers), and execution phase (controlling vehicle). This segmentation allows each phase to be handled by specialized modules, reducing overall system complexity while maintaining high reliability through focused optimization of each segment.
Solution Approach 2:
The system performs preliminary actions by pre-calculating velocity obstacles and potential collision paths before actual collision risk materializes. By anticipating possible collision scenarios and pre-determining avoidance maneuvers, the system reduces real-time computational burden while ensuring reliable collision avoidance when needed.
2Reliability
If real-time determination of optimal maneuvers for multiple obstacles is implemented, then collision avoidance effectiveness improves, but computational load and processing time increase
Solution Approach 1:
The patent replaces complex mechanical computation with mathematical modeling of velocity obstacles. By using analytical solutions and geometric relationships rather than iterative simulation, the system achieves real-time determination of optimal maneuvers for multiple obstacles without excessive computational load, maintaining both effectiveness and speed.
3Loss of time
If escalating severity warnings are provided to drivers, then driver awareness and response time improve, but information processing requirements increase
Solution Approach 1:
The patent applies local quality by providing differentiated warning levels tailored to specific collision risk scenarios. Instead of uniform information delivery, the system analyzes the local characteristics of each detected obstacle and potential collision path, then provides appropriately scaled warnings that match the severity and immediacy of the threat, optimizing driver response while minimizing unnecessary information processing.
4Object-affected harmful factors
If safety accessories like airbags are activated when collision is unavoidable, then impact mitigation improves, but false activation risks increase
Solution Approach 1:
The patent implements preliminary anti-action by preparing safety systems in advance but with conditional activation criteria. The system pre-positions safety accessories and establishes clear thresholds for unavoidable collision determination, activating protective measures only when analysis confirms that collision cannot be avoided through any feasible maneuver, thereby minimizing false activation while maximizing protection when truly needed.
Data Source
AI summary
Method for transmitting a warning signal to a driver of a driven vehicle regarding an impending collision with a moving and/or stationary object in the vicinity of the driven vehicle. The method comprises the following steps of providing the driven vehicle with means for obtaining updated data regarding, position, velocity vector and predicted moving path of the objects; selecting a series of one or more time horizons having decreasing or increasing duration; for the longest of the selected time horizons: generating a linear velocity object (LVO) and/or non-linear velocity object (NLVO) of each of the objects; selecting a sampling time interval Δt, during which an LVO and/or NLVO is generated; determining a range of feasible velocity vector changes for the driven vehicle that are attainable within a performance time interval ΔT; repeatedly providing the driver, after each Δt, with information regarding feasible velocity vector changes for the performance time interval; sensing, estimating or assuming dynamic changes parameters representing the movement of the driven vehicle within the performance time interval, and whenever required, generating a warning signal with an escalating severity level that reflects the relative imminence of collision with the objects and that corresponds to the longest time horizon; repeating the steps above, while each time generating an updated LVO and/or NLVO for a subsequent sampling time interval, until reaching another selected time horizon which is shorter than a previously selected time horizon and another selected time horizon, until collision is unavoidable.