Autonomous Vehicle Trajectory Planning With Dynamic Risk Budgets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous vehicle planning systems face challenges in balancing safety and efficiency, often behaving too conservatively to meet high safety thresholds, which can lead to suboptimal performance and require close human supervision, while non-conservative approaches may not provide sufficient safety guarantees.
Innovation Solution
The implementation of a risk-budgeting system that iteratively plans and adjusts trajectories based on a risk budget, allowing for feasible actions while ensuring safety constraints are met, including the use of emergency stops to maintain passive safety in case of collisions, and dynamically reallocating risk budgets across objectives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If autonomous vehicle planning systems adopt conservative approaches to meet high safety thresholds, then safety reliability is improved, but operational efficiency and performance deteriorate
Solution Approach 1:
The system dynamically adjusts the risk budget parameter to control the degree of conservatism in planning. By modifying this parameter, the system can transition between conservative and non-conservative planning modes, thereby balancing safety reliability with operational efficiency based on current driving conditions and risk assessments
Solution Approach 2:
The planning system transitions from static conservative planning to dynamic non-conservative planning with adaptive risk management. The system continuously updates the risk budget and replans trajectories in real-time, allowing it to adapt to changing environmental conditions while maintaining safety guarantees through probabilistic risk bounds
2Productivity
If autonomous vehicle planning systems adopt non-conservative approaches to improve operational efficiency, then productivity is improved, but safety guarantees deteriorate
Solution Approach 1:
The system implements continuous feedback through risk assessment and monitoring mechanisms. By evaluating the actual risk incurred during execution and comparing it against the allocated risk budget, the system can adjust future planning decisions to ensure safety guarantees are maintained while allowing non-conservative actions when appropriate
Solution Approach 2:
The system performs preliminary risk assessment and budget allocation before executing planned actions. By pre-calculating the risk costs of potential actions and allocating appropriate risk budgets in advance, the system enables non-conservative planning while maintaining safety guarantees through proactive risk management
3Reliability
If autonomous vehicle planning systems require close human supervision to ensure safety, then safety reliability is improved, but device complexity and operational ease deteriorate
Solution Approach 1:
The system implements self-service through autonomous risk management and self-correction capabilities. By automatically assessing risks, adjusting planning strategies, and correcting deviations from safe operation without human intervention, the system maintains high safety reliability while reducing the complexity of human supervision requirements
4Reliability
If autonomous vehicle planning systems allocate fixed risk budgets to trajectories, then safety guarantees are improved, but adaptability to changing conditions deteriorates
Solution Approach 1:
The system employs dynamic risk budget allocation that adapts to changing driving conditions, environmental factors, and vehicle states. By continuously updating risk budgets based on real-time assessments rather than using fixed allocations, the system maintains safety guarantees while improving adaptability to varying operational contexts
Data Source
AI summary
Techniques for safe non-conservative planning include: obtaining a risk budget constraining a plan for an autonomous vehicle to satisfy an objective; based at least on the risk budget and the objective, planning a trajectory of the autonomous vehicle toward the objective, at least by: (a) determining a risk cost associated with an initial planned action of the trajectory, (b) based at least on the risk cost, determining whether the trajectory is feasible or infeasible within the risk budget, and (c) responsive to determining that the trajectory is feasible within the risk budget, executing the initial planned action; decreasing the risk budget by the risk cost, to obtain a remaining risk budget; obtaining state data corresponding to a state of the autonomous vehicle after executing the initial planned action; and based at least on the state data, the remaining risk budget, and the objective, planning another trajectory toward the objective.