Obstacle-Point Velocity Constraints for Vehicle Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Semi-autonomous and robotic vehicles face challenges in obstacle avoidance, as they rely on automated techniques lacking the real-time navigational decisions humans can make to avoid collisions.
Innovation Solution
A processor-based method for vehicles to receive sensor data, determine distance and direction to obstacles, and adjust velocity to avoid collisions by determining velocity constraints and selecting appropriate velocity solutions that match navigation instructions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automated obstacle avoidance techniques are used in semi-autonomous and robotic vehicles, then collision avoidance capability is improved, but the system complexity and computational requirements increase
Solution Approach 1:
The patent segments the obstacle avoidance problem into discrete velocity constraints for each detected obstacle point. The sensor data is processed to identify individual points representing obstacles, and velocity constraints are calculated independently for each point based on its distance and direction from the vehicle. This segmentation allows the complex automated avoidance system to break down the continuous navigation problem into manageable discrete constraints that can be efficiently processed.
Solution Approach 2:
The patent transforms the obstacle avoidance problem by changing parameters from spatial coordinates to velocity constraints. For each obstacle point, the system calculates distance and direction parameters, then converts these into velocity inequality constraints that directly modify the vehicle's commanded velocity. This parameter transformation simplifies the computational burden by working directly with velocity adjustments rather than complex trajectory replanning.
2Measurement precision
If velocity constraints are determined for each obstacle point based on distance and direction, then collision avoidance precision is improved, but computational burden increases
Solution Approach 1:
The patent applies partial action by focusing computational resources only on the necessary velocity adjustments for each obstacle point rather than recalculating the entire navigation path. The system determines velocity constraints for each detected point and applies only the necessary modifications to the commanded velocity, avoiding excessive computation of unrelated navigation parameters. This selective approach maintains precision while improving computational efficiency.
Solution Approach 2:
The patent replaces complex mechanical navigation decision-making with mathematical velocity constraint calculations. Instead of using intricate path planning algorithms or human-like navigational reasoning, the system substitutes these with direct mathematical computations of velocity inequalities based on obstacle distance and direction. This substitution dramatically reduces computational burden while maintaining avoidance precision.
3Reliability
If the vehicle adjusts velocity to satisfy multiple velocity constraints, then safety is improved, but the ability to follow navigation instructions may be compromised
Solution Approach 1:
The patent implements feedback by continuously monitoring the relationship between commanded velocity, velocity constraints, and actual vehicle velocity. The system determines whether the commanded velocity satisfies all velocity constraints, and if not, it adjusts the velocity by subtracting the obstacle-directed component. This feedback loop ensures that safety constraints are continuously enforced while maintaining alignment with navigation instructions as much as possible. The feedback mechanism allows the system to dynamically balance safety requirements with navigation instruction following.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments include devices and methods for vehicle collision avoidance. A processor of the vehicle may receive sensor data comprising one or more points. The processor may determine a distance and a direction from the vehicle to the one or more points. The processor may determine a velocity constraint for each of the one or more points based on the determined distance and direction from the vehicle to the one or more points. Based on a navigation instruction, the processor may determine one or more velocity solutions that satisfy the one or more velocity constraints. The processor may select a velocity solution from the determined one or more velocity solutions based on the navigation instruction, and may adjust a vehicle velocity based on the selected velocity solution.