Vehicle Lateral Guidance Obstacle Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for lateral vehicle guidance during evasive maneuvers are limited in their ability to differentiate between traversable and non-traversable obstacles, particularly at varying speeds and angles, which can lead to inefficient collision avoidance and reduced driving comfort.
Innovation Solution
A method that classifies obstacles based on their height using an object detection sensor, allowing vehicles to drive over low-height obstacles when no collision-free trajectory is available, and adjusts steering signals based on criticality and vehicle speed to adapt to driving situations, enabling adaptive system behavior.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If obstacles are classified as non-traversable to ensure collision avoidance, then safety is improved, but driving comfort and maneuverability deteriorate due to inability to drive over low-height obstacles during normal maneuvers
Solution Approach 1:
The system dynamically changes the height parameter threshold for obstacle classification based on vehicle speed. At low speeds, higher thresholds allow driving over curbs and low obstacles during parking maneuvers. At high speeds, lower thresholds prevent driving over obstacles that could endanger pedestrians or cause vehicle damage, thus resolving the contradiction between safety and driving comfort through parameter adaptation.
Solution Approach 2:
The obstacle traversability classification is made dynamic rather than static. The system continuously adapts the classification criteria based on real-time driving conditions, particularly vehicle speed and angle of approach. This allows the same physical obstacle (e.g., a curb) to be classified as traversable during low-speed parking but non-traversable during high-speed emergency maneuvers, resolving the safety-comfort contradiction.
2Reliability
If all obstacles are treated as non-traversable, then collision avoidance is maximized, but system adaptability to different driving situations deteriorates
Solution Approach 1:
The system employs multiple height threshold parameters that are selectively applied based on driving situation. Instead of a single static threshold, the system uses speed-dependent parameters that automatically adjust the obstacle classification criteria, enabling the system to adapt to different driving scenarios while maintaining collision avoidance safety.
Solution Approach 2:
The obstacle classification system is made dynamic by continuously monitoring driving conditions and adjusting classification parameters in real-time. The system transitions between different classification modes (permissive at low speed, restrictive at high speed) based on the driving situation, thereby achieving both high reliability and adaptability.
3Manufacturing precision
If low-height obstacles are masked out to enable collision-free trajectory calculation, then trajectory planning accuracy is improved, but the ability to drive over traversable obstacles deteriorates
Solution Approach 1:
The system dynamically adjusts the masking parameter for low-height obstacles based on vehicle speed and trajectory criticality. During low-speed maneuvers with low criticality, the masking threshold is raised to allow trajectory planning that drives over curbs and low obstacles. During high-speed emergency maneuvers, the masking threshold is lowered to ensure collision-free trajectories, thus resolving the contradiction between planning accuracy and adaptability.
Solution Approach 2:
The obstacle masking strategy is made dynamic rather than static. The system continuously evaluates trajectory criticality and adjusts which obstacles are masked during path planning. This allows the same obstacle to be masked in critical emergency situations but unmasked during non-critical normal maneuvers, achieving both accurate trajectory planning and operational versatility.
4Ease of operation
If the system allows driving over low-height obstacles at low speeds, then driving comfort is improved, but the risk of affecting pedestrians or causing vehicle damage increases at higher speeds
Solution Approach 1:
The system changes the height threshold parameter for obstacle traversability based on vehicle speed. At low speeds, higher thresholds permit driving over curbs and low obstacles for comfort. At high speeds, lower thresholds prevent such actions that could harm pedestrians or damage the vehicle, thus resolving the contradiction between comfort and safety risks through speed-dependent parameter adaptation.
Solution Approach 2:
The obstacle traversability classification is dynamically adjusted based on real-time speed measurement. The system transitions from a permissive classification mode at low speeds to a restrictive mode at high speeds, automatically adapting the same physical system to different operating conditions to balance comfort and harm prevention.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and device for the lateral guidance of a motor vehicle (4), in particular for providing evasive assistance to a motor vehicle, is proposed, with which a collision can be avoided. An object detection sensor (5) detects obstacles (3, 7, 8, 9) and determines information regarding their position relative to the vehicle and at least one other property. The obstacles are then classified as "crossable" or "not crossable." Furthermore, a trajectory is determined that describes a collision-free driving corridor away from the detected obstacles. Steering signals are determined according to the trajectory and output to a lateral controller for evasive assistance or automatic evasive maneuvering of the vehicle. The classification as "crossable" or "not crossable" is performed based on the height of the obstacle as determined by the object detection sensor.