Evasive Steering Path Planning Using Fifth-Order Polynomials
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle path generation systems do not provide effective collision avoidance in semi-autonomous or autonomously driven vehicles when a moving object is detected, as they fail to account for objects moving transversely or longitudinally relative to the vehicle's path.
Innovation Solution
A system and method that determine a predicted vehicle path, detect moving objects, and calculate a steering correction path around the object using fifth-order polynomial equations to avoid collisions, including a collision avoidance path, a cruise path, and a return path, depending on the object's direction and proximity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle follows the predicted path based on current vehicle state, then the vehicle maintains smooth lane centering and lane changing control, but the vehicle may collide with moving objects detected in front
Solution Approach 1:
The system performs preliminary detection of moving objects and calculates evasive paths in advance before collision occurs. The path prediction processor continuously predicts potential collision scenarios and pre-computes evasive maneuvers using fifth-order polynomial equations, allowing the vehicle to execute smooth path adjustments when objects are detected rather than reacting abruptly to immediate threats.
2Reliability
If the system implements evasive steering maneuvers to avoid collisions with moving objects, then collision avoidance is improved, but the vehicle path generation becomes more complex with multiple path segments
Solution Approach 1:
The evasive path is segmented into three distinct polynomial path segments: collision avoidance path (to bypass the object), cruise path (to maintain safe distance while passing), and return path (to return to original trajectory). Each segment is calculated using fifth-order polynomial equations with specific boundary conditions, allowing the complex evasive maneuver to be broken down into manageable, smoothly transitioned segments that can be executed sequentially.
3Reliability
If the vehicle automatically steers along the collision avoidance path calculated using fifth-order polynomial equations, then collision avoidance is achieved, but the maneuver may affect vehicle stability and passenger comfort
Solution Approach 1:
The path prediction processor dynamically adjusts the evasive path parameters based on real-time vehicle state information including longitudinal speed, lateral speed, and yaw-rate. The fifth-order polynomial equations are solved with boundary conditions that account for current vehicle dynamics, ensuring the evasive maneuver adapts to the vehicle's actual motion state rather than applying fixed geometric paths, thereby maintaining stability during the maneuver.
Data Source
AI summary
A system and method for providing path planning and generation in a manually driven vehicle that provides a steering correction for collision avoidance purposes. The method includes determining a predicted path of the vehicle based on vehicle motion sensing data and vehicle parameters. The method also includes detecting a moving object in front of the vehicle and determining if a collision between the vehicle and the object will occur if the vehicle travels along the predicted path at the current vehicle speed. The method solves a fifth-order polynomial equation to define a collision avoidance path from the current vehicle position to a waypoint a safe distance from the object and a return path from the waypoint to the lane center that the vehicle is automatically steered along.