Autonomous Vehicle Trajectory Planning with Dynamic Lane Biasing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating routes for autonomous vehicles are computationally intensive and may not provide safe or comfortable routes for passengers, as they do not effectively adapt to varying environmental conditions such as object classification and vehicle velocity.
Innovation Solution
The techniques involve adaptively scaling the density of trajectory points based on cost-associated factors like curvature and obstacle proximity, using higher densities in high-activity areas and lower densities in low-activity areas, and dynamically adjusting buffer sizes around objects based on classification and velocity to optimize route planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If uniform high density of trajectory points is used throughout the route, then route accuracy and safety are improved, but computational burden increases significantly
Solution Approach 1:
The patent applies local quality by varying the density of trajectory points based on the characteristics of different route segments. High-density sampling is applied to high-activity areas (curved paths, areas with obstacles, intersections) while low-density sampling is used for straight, clear paths. This resolves the contradiction by providing high accuracy only where needed rather than uniformly across the entire route.
Solution Approach 2:
The system dynamically adjusts the density of trajectory points based on real-time environmental factors such as curvature, obstacle proximity, and activity level. The density is not fixed but adapts to the specific conditions of each route segment, allowing the system to maintain high accuracy in critical areas while reducing computational load in safer areas.
2Reliability
If fixed buffer size is used around objects, then system complexity is reduced, but safety and comfort are compromised as it cannot adapt to varying environmental conditions
Solution Approach 1:
The patent changes the buffer size parameter dynamically based on object classification (pedestrian, vehicle, cyclist, animal) and environmental conditions (vehicle velocity, curvature). Different object types receive different buffer sizes, and these sizes are further adjusted according to the vehicle's speed and the route's curvature, providing adaptive safety margins without requiring overly complex decision logic.
Solution Approach 2:
The buffer size is made dynamic rather than fixed, allowing it to expand or contract based on real-time conditions. Faster vehicle speeds result in larger buffers, while slower speeds allow for smaller buffers. This dynamic adjustment maintains safety and comfort across varying operating conditions without requiring the system to be maximally complex.
3Speed
If route planning does not consider object classification and velocity, then processing speed is maintained, but passenger comfort and safety are reduced
Solution Approach 1:
The system applies different processing priorities and buffer sizes to different object types based on their classification. Vulnerable road users (pedestrians, cyclists, animals) receive larger buffers and higher processing priority, while other vehicles receive standard treatment. This local differentiation improves comfort and safety for vulnerable users without requiring uniform high-processing levels for all objects.
Solution Approach 2:
The system changes key parameters (buffer size, trajectory point density, cost function weights) based on object classification and velocity. This allows the processing to be optimized for each situation - using larger buffers and higher density only when needed for safety-critical objects, rather than applying maximum processing resources to all situations uniformly.
Data Source
AI summary
Techniques for generating trajectories and drivable areas for navigating a vehicle in an environment are discussed herein. The techniques can include receiving a reference trajectory representing an initial trajectory for a vehicle, such as an autonomous vehicle, to traverse the environment in a first drivable area. An object within a distance threshold can be detected in the environment and a second drivable area can be determined. Further, the techniques can include determining a target trajectory based at least in part on the reference trajectory and/or the second drivable area which can provide a region for the object to traverse the environment, and controlling the autonomous vehicle to traverse the environment based at least in part on the target trajectory.