Autonomous Obstacle Mapping for Unmapped Road Hazard Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving systems face challenges in effectively detecting and avoiding obstacles, particularly those not mapped on roadways, such as potholes, due to limitations in sensor detection and reaction time.
Innovation Solution
An autonomous driving system comprising an I/O module, sensors, processing circuitry, and an autonomous driving manager that detects obstacles using sensor data and sends reports to an obstacle avoidance server, employing a signature generation method that reduces power consumption by iteratively expanding and merging dimensions to enhance obstacle detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional sensor-based obstacle detection is used, then the system can detect obstacles, but the detection accuracy is insufficient for unmapped obstacles like potholes
Solution Approach 1:
The patent transitions from traditional 2D camera-based detection to 3D point cloud data from LiDAR sensors, adding spatial dimensionality to obstacle detection. This enables accurate detection of unmapped obstacles like potholes by capturing three-dimensional spatial information, directly improving detection accuracy for previously undetectable obstacle types.
Solution Approach 2:
The patent introduces a server-based obstacle database as an intermediary between sensors and the autonomous driving manager. The server stores and processes point cloud data to create detailed obstacle models, acting as a mediator that enhances detection capabilities by providing pre-processed obstacle information to the vehicle system.
2Measurement precision
If comprehensive sensor data processing is performed to improve obstacle detection, then detection accuracy improves, but power consumption increases
Solution Approach 1:
The patent segments the obstacle detection process into two parts: real-time sensor data collection in the vehicle and computationally intensive point cloud processing on a remote server. This segmentation allows the vehicle to maintain low power consumption while achieving high detection accuracy through server-based processing of the segmented and analyzed point cloud data.
Solution Approach 2:
The server acts as an intermediary that performs the computationally intensive point cloud processing, freeing the vehicle's onboard systems from heavy computational loads. This reduces the vehicle's power consumption while maintaining high obstacle detection accuracy through the server's processing capabilities.
Data Source
AI summary
An autonomous driving system for a vehicle comprising: an I/O module operative to communicate with an obstacle avoidance server; at least one sensor operative to provide at least an indication of an obstacle in a path of the vehicle; processing circuitry; and an autonomous driving manager to be executed by the processing circuitry and operative to: detect the at least an indication of an obstacle based on data provided by the at least one sensor, drive the vehicle in accordance with a driving policy associated with the obstacle, and send an obstacle report with obstacle information associated with the at least an indication of an obstacle to the obstacle avoidance server.