Autonomous Vehicle Speed Planning via Station-Time Graph Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles face challenges in determining optimal speeds along a path, especially when navigating through complex environments with obstacles and varying traffic conditions, which can lead to unstable vehicle control and increased risk of accidents.
Innovation Solution
A method utilizing a station-time graph with applied kernels and constraints, such as driving, history, and dragging kernels, along with initial, vehicle, speed limit, and curvature constraints, to determine optimal speeds through quadratic programming optimization, ensuring stable and safe navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional speed determination methods are used in complex environments, then the vehicle can navigate basic paths, but the vehicle control becomes unstable and accident risk increases
Solution Approach 1:
The speed planning problem is segmented into multiple discrete speed values at different time steps, represented as variables in a quadratic programming formulation. This allows the complex continuous optimization problem to be broken down into manageable discrete decisions that can be solved efficiently while ensuring control stability.
Solution Approach 2:
The patent transforms the speed planning problem from a simple temporal sequence into a multi-dimensional optimization space by incorporating spatial position, velocity, acceleration, and constraint satisfaction dimensions. This dimensional expansion allows simultaneous optimization of multiple control aspects, improving reliability while managing complexity through structured mathematical formulation.
2Adaptability or versatility
If speed adjustments are made frequently to respond to environmental changes, then navigation adaptability improves, but vehicle control stability deteriorates
Solution Approach 1:
The system dynamically adjusts speed plans by formulating the optimization problem at each time step with updated environmental constraints and vehicle states. The quadratic programming solver dynamically computes optimal speed adjustments that adapt to changing conditions while maintaining smooth transitions through continuous optimization variables representing velocity and acceleration.
Solution Approach 2:
The patent changes the parameter representation from direct speed commands to optimized velocity and acceleration variables subject to constraints. By parameterizing the speed trajectory with physical constraints on maximum acceleration and velocity limits, the system achieves adaptability to environmental changes while ensuring smooth, stable transitions through mathematically bounded parameter variations.
3Reliability
If conservative speed limits are applied to ensure safety, then accident risk decreases, but navigation efficiency and productivity reduce
Solution Approach 1:
The system optimizes safety and efficiency by changing from fixed conservative speed limits to dynamic speed profiles computed through quadratic programming. The optimization incorporates safety constraints (maximum acceleration, velocity bounds, obstacle avoidance) while maximizing navigation efficiency by selecting the highest feasible speeds at each time step, achieving both safety and productivity simultaneously.
Solution Approach 2:
The patent applies partial constraint satisfaction by not enforcing maximum safety margins at all times, but rather applying constraints selectively based on environmental conditions. The quadratic programming formulation allows the vehicle to operate at higher speeds when conditions permit, applying safety constraints only when necessary, thus achieving safety without excessive conservatism that would reduce navigation efficiency.
Data Source
AI summary
A station-time (S-T) graph may be obtained in response to a first reference line representing a path from a first location to a second location associated with an autonomous driving vehicle (ADV). One or more kernels may be applied to the S-T graph. Each of the one or more kernels may indicate a plurality of points on the S-T graph. One or more constraints may be applied to the S-T graph. Each of the one or more constraints may indicate a condition for points in the S-T graph. A set of speeds for portions of the path is determined based on the one or more kernels and the one or more constraints.