Autonomous Driving Speed Control With Right-of-Way-Aware Obstacle Trajectories
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous vehicle systems face irrational speed planning due to inconsistencies between predicted and actual obstacle trajectories, leading to unreasonable braking and decreased driving rationality.
Innovation Solution
Incorporating right-of-way information into obstacle trajectory predictions to determine target speed information, ensuring the vehicle travels according to a planned route and speed that avoids collisions and unnecessary deceleration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle brakes to a stop to avoid the obstacle based on predicted trajectory, then collision avoidance is achieved, but driving rationality deteriorates due to unnecessary braking when the obstacle actually yields
Solution Approach 1:
The system establishes a feedback mechanism by continuously monitoring the actual trajectories of surrounding vehicles and comparing them with predicted trajectories. When a vehicle is predicted to cut in but actually maintains its lane (yielding to the ego vehicle), the system receives feedback that the obstacle will not collide, allowing the ego vehicle to resume normal speed without unnecessary braking. This feedback loop resolves the contradiction by enabling the system to distinguish between actual and potential collision risks.
Solution Approach 2:
The system performs preliminary trajectory prediction and right-of-way determination before actual collision occurs. By predicting obstacle trajectories and determining right-of-way status in advance, the system can proactively adjust speed planning. When the prediction indicates the obstacle will yield (based on right-of-way rules), the system takes preliminary action to maintain normal speed rather than braking, thus avoiding unnecessary deceleration while still ensuring collision avoidance.
2Reliability
If the vehicle plans speed based on predicted obstacle trajectory, then collision risk is reduced, but driving rationality worsens due to inconsistency between predicted and actual trajectories
Solution Approach 1:
The system uses feedback from actual obstacle behavior to correct speed planning. When the actual trajectory of an obstacle differs from the predicted trajectory (e.g., the obstacle yields when it was predicted to cut in), the system receives feedback that updates the speed planning. This allows the vehicle to maintain rational driving behavior by adjusting speed based on actual rather than purely predicted trajectories.
Solution Approach 2:
The system changes the parameters used in speed planning by incorporating right-of-way information and actual trajectory data alongside predicted trajectories. By modifying the input parameters from solely predicted trajectory to a combination of predicted trajectory, actual trajectory, and right-of-way status, the system achieves more rational speed planning that reflects real-world driving conditions while maintaining collision avoidance.
3Reliability
If the vehicle decelerates to ensure safety based on obstacle prediction, then safety is improved, but driving performance worsens due to unnecessary deceleration
Solution Approach 1:
The system employs feedback mechanisms to monitor whether deceleration is actually necessary. When the ego vehicle decelerates based on predicted obstacle trajectory but the obstacle actually yields (as detected by trajectory monitoring), the feedback indicates that deceleration was unnecessary. The system uses this feedback to adjust future speed planning, reducing unnecessary deceleration events while maintaining safety when actual collision risk is present.
Solution Approach 2:
The system performs preliminary assessment of collision risk by analyzing right-of-way information and actual obstacle trajectories before executing deceleration. When the preliminary assessment indicates that the obstacle will yield (based on right-of-way rules and actual behavior patterns), the system takes preliminary action to maintain normal speed. This prevents unnecessary deceleration while still ensuring safety when actual collision risk is identified.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present application provides a vehicle driving control method and apparatus, relates to the field of autonomous driving in computer technology. The specific implementation scheme is: determining, according to driving data of a vehicle, at least one piece of candidate speed information of the vehicle, where the candidate speed information is used to indicate traveling speed of the vehicle at each moment in a first preset time period; predicting, according to movement data of an obstacle on a road, an obstacle trajectory of the obstacle in the first preset time period; determining right-of-way information corresponding to the obstacle trajectory, where the right-of-way information is used to indicate a priority in traveling of both the vehicle and the obstacle; and determining target speed information in the at least one piece of candidate speed information according to both the right-of-way information corresponding to the obstacle trajectory and the obstacle trajectory, and controlling the vehicle to travel according to both a route corresponding to the vehicle and the target speed information. The target speed information is selected according to the right-of-way information, so that unnecessary deceleration of the vehicle can be avoided, which effectively improves the driving rationality of the vehicle.