Autonomous Vehicle Trajectory Replanning for Emergency Obstacle Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in emergency obstacle avoidance due to the limitations of existing trajectory replanning methods, which often result in instability and computational inefficiency, particularly when offline reference trajectories become infeasible or are influenced by changing environments.
Innovation Solution
A method for iterative trajectory replanning that updates reference trajectories in real-time, incorporating nonlinear vehicle dynamics, such as tire dynamics, to maintain stability and efficiency, using a combination of model predictive control (MPC) and spatial iterative replanning (SpIRe) to adapt to sudden obstacles and environmental changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If offline reference trajectories are used for trajectory planning, then computational efficiency is improved, but trajectory model fidelity deteriorates when environments change
Solution Approach 1:
The system dynamically switches between offline reference trajectories and online optimal sequences based on environmental conditions. When obstacles are detected, the system transitions from using pre-computed offline trajectories to generating real-time optimal sequences, ensuring both computational efficiency in normal conditions and trajectory fidelity in changing environments.
Solution Approach 2:
The system continuously monitors environmental changes and vehicle state, using this feedback to determine when to abandon offline reference trajectories and switch to online optimal sequence generation. This feedback mechanism ensures trajectory model fidelity is maintained when environments change while preserving computational efficiency when conditions are stable.
2Reliability
If completely new trajectories are computed in real-time, then trajectory model fidelity is improved, but computational efficiency deteriorates
Solution Approach 1:
Offline reference trajectories are pre-computed and stored before real-time operation. When environmental conditions permit, these pre-computed trajectories are used directly, avoiding the computational burden of real-time optimization. This preliminary action ensures trajectory model fidelity is available when needed while preserving computational efficiency during normal operation.
Solution Approach 2:
Instead of computing completely new trajectories in all scenarios, the system computes optimal sequences only partially - specifically when environmental changes or obstacles require it. This partial action approach maintains trajectory fidelity in critical situations while avoiding the excessive computational cost of full real-time recomputation in all cases.
3Productivity
If offline reference trajectories are used, then computational efficiency is improved, but adaptability to changing environments deteriorates
Solution Approach 1:
The system uses continuous feedback from environmental sensors to detect changes in the operating environment. When changes are detected, the feedback triggers a switch from offline reference trajectories to online optimal sequence generation, ensuring adaptability to changing environments while maintaining computational efficiency when conditions remain stable.
Solution Approach 2:
The trajectory planning system is dynamic, automatically adjusting its approach based on environmental conditions. It transitions from static offline reference trajectories to dynamic online optimal sequences when adaptability is required, providing both computational efficiency and environmental adaptability as needed.
4Adaptability or versatility
If online optimal sequences are computed in real-time, then adaptability to changing environments is improved, but computational efficiency deteriorates
Solution Approach 1:
Offline reference trajectories are pre-computed and stored as a foundation for real-time planning. This preliminary action provides a computationally efficient baseline that can be used directly when environments are stable, reducing the need for expensive real-time optimization and improving overall computational efficiency while maintaining adaptability when needed.
Data Source
AI summary
Systems and methods of trajectory planning for an autonomous vehicle are disclosed. Exemplary implementations may: determine a first trajectory plan for the vehicle traveling along a first spatial location at a first point in time, the first trajectory plan being a reference trajectory plan; compute an optimal sequence for the vehicle traveling along a second spatial location at a second point in time subsequent the first point in time; and calculate a second trajectory plan for the vehicle by updating the first trajectory plan with information from the computed optimal sequence.