Autonomous Vehicle Deceleration Planning for Road Object Traversal
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicle planning algorithms often fail to safely navigate sudden objects on the road, either risking rear-end collisions or excessive damage by swerving or stopping, as they lack effective deceleration strategies for passing over objects.
Innovation Solution
A system and method for generating a vehicle deceleration plan that determines the safest trajectory to pass over an object by adjusting deceleration rates based on object size, vehicle clearance, and preceding vehicle reaction time, balancing the risks of rear-end collisions and object damage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If the vehicle comes to a complete stop or swerves to avoid the object, then the risk of damage from hitting the object is reduced, but the risk of rear-end collision with the preceding vehicle increases
Solution Approach 1:
The system dynamically adjusts the deceleration rate based on real-time conditions including distance to object, speed of vehicle, and distance to preceding vehicle. Rather than using fixed stop or swerve commands, the planner continuously optimizes the deceleration profile to balance avoiding object damage while preventing rear-end collisions
Solution Approach 2:
The system changes the parameter of deceleration rate from static (complete stop) to variable (controlled deceleration). By adjusting the deceleration rate as a continuous parameter based on sensor data and environmental conditions, the system finds an optimal intermediate state between stopping completely and maintaining speed, thereby reducing both object damage risk and rear-end collision risk
2Object-affected harmful factors
If the vehicle decelerates suddenly to avoid the object, then the risk of damage from hitting the object is reduced, but the preceding vehicle may not have time to react and could collide with the subject vehicle
Solution Approach 1:
The system performs preliminary assessment of the situation by calculating the time gap between the subject vehicle and preceding vehicle before initiating deceleration. This preliminary action allows the system to determine whether sudden deceleration is safe or if a more gradual approach is needed to give the preceding vehicle adequate reaction time
Solution Approach 2:
The system uses feedback from sensors monitoring the preceding vehicle's position and speed to continuously adjust the deceleration strategy. By incorporating real-time feedback about the preceding vehicle's state, the system can modify its deceleration profile to ensure the preceding vehicle has sufficient time to react and avoid collision
3Reliability
If the vehicle maintains speed to avoid rear-end collision, then the risk of rear-end collision is reduced, but the risk of damage from hitting the object increases
Solution Approach 1:
The system changes the speed parameter from constant (maintaining speed) to variable (controlled deceleration). By adjusting speed as a dynamic parameter based on the relative risks of hitting the object versus causing a rear-end collision, the system optimizes the trade-off between these two harmful outcomes
Data Source
AI summary
Methods and systems address a situation in which an autonomous vehicle encounters an object while driving. A system may include determining a vehicle trajectory of a subject vehicle passing over the object detected in front of the subject vehicle, and in response to determining the vehicle trajectory, generating a vehicle deceleration plan for approaching the object and traveling over the object. The system may include executing the vehicle deceleration plan as the subject vehicle travels over the object, where the vehicle deceleration plan is based on a preceding vehicle located behind the subject vehicle and a reaction time of the preceding vehicle as it reacts to deceleration of the subject vehicle.