Autonomous Vehicle Trajectory Planning for Vegetation Misclassification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face difficulties in distinguishing between different types of obstacles, particularly vegetation, using LiDAR sensors, which can lead to misclassification and inefficient navigation, including over-braking and potential rear-end collisions.
Innovation Solution
A method and system that combines LiDAR and camera data to generate point clouds and images, perform factor queries, and label obstacles as collidable or non-collidable, enabling the vehicle to adjust its trajectory and actions accordingly, such as speed and path planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LiDAR is used to detect obstacles, then the presence and absence of objects can be detected, but vegetation is misclassified as vehicles leading to over-braking
Solution Approach 1:
The patent combines LiDAR point cloud data with camera image data to create a multi-sensor fusion system. The LiDAR provides accurate depth and spatial information while the camera provides visual texture and color information, allowing the system to distinguish vegetation from vehicles more reliably than either sensor alone.
Solution Approach 2:
The system utilizes color information from camera images to differentiate between vegetation and vehicles. Vegetation typically exhibits green color characteristics while vehicles have diverse colors, allowing the classification algorithm to use color as a discriminative feature to reduce misclassification.
2Reliability
If the vehicle stops for vegetation to avoid collision, then safety is improved, but ride quality degrades and rear-end collisions increase
Solution Approach 1:
The system applies different collision avoidance strategies based on the local characteristics of detected objects. For vegetation, the system determines that collision is acceptable and adjusts behavior accordingly, while for actual vehicles or pedestrians, strict avoidance is maintained. This localized differentiation improves ride quality while maintaining safety.
3Adaptability or versatility
If LiDAR classifies all objects uniformly, then detection coverage is maximized, but differentiation between object types becomes impossible
Solution Approach 1:
The patent adds a new dimension of information by incorporating camera imagery alongside LiDAR point cloud data. This multi-dimensional approach allows the system to maintain comprehensive detection coverage from LiDAR while adding visual characteristics from the camera to enable precise object type differentiation.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Improves the accuracy of obstacle classification, reducing misclassification of vegetation and enhancing navigation efficiency by allowing the vehicle to differentiate between collidable and non-collidable objects, thereby preventing unnecessary braking and improving safety.
Implementation Method 1
A LIDAR sensor is configured to emit light, which strikes material (e.g., objects) within the vicinity of the LiDAR sensor. Once the light contacts the material, the light is deflected. Some of the deflected light bounces back to the LiDAR sensor.
Data Source
AI summary
Systems and methods for planning a trajectory of a vehicle based on one or more obstacles is provided. The method may comprise generating one or more data points from one or more sensors coupled to a vehicle, and, using a processor, detecting one or more obstacles within a LiDAR point cloud, generating a patch for each of the one or more obstacles, projecting the LiDAR point cloud into the image, wherein each patch represents a region of an image for each of the one or more obstacles, performing a factor query on the image for each of the one or more obstacles, for each of the one or more obstacles, based on the factor query, determining a label for the obstacle and labeling the obstacle with the label, and planning a trajectory of the vehicle. The label may indicate a collidability of the obstacle.