Autonomous Driving Path Generation Using Stochastic Trajectory Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving systems face limitations in precision due to systematic constraints of vehicle-mounted sensors and inaccuracies caused by outdated communication information, leading to suboptimal collision avoidance and path management.
Innovation Solution
An autonomous driving device that incorporates a sensor to gather movement data of surrounding objects, a controller to generate paths, and a surrounding object analyzer to stochastically analyze expected trajectories using big data, enhancing path accuracy and collision avoidance by updating vehicle paths in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If autonomous driving is performed using only vehicle-mounted sensor data, then the system can operate independently, but the preciseness of autonomous driving control decreases due to systematic sensor limits
Solution Approach 1:
The patent combines vehicle-mounted sensor data with map data from communication systems to create a more accurate and comprehensive representation of the surrounding environment. This merging of multiple data sources compensates for the systematic limitations of individual sensors and improves overall measurement precision for autonomous driving control.
Solution Approach 2:
The patent introduces map data as an intermediary information source that mediates between the vehicle's sensor data and the autonomous driving control system. This intermediary provides additional contextual information about the environment that enhances the precision of path generation and collision avoidance decisions.
2Loss of information
If autonomous driving path is obtained through communication systems, then path information can be acquired, but the accuracy decreases due to non-up-to-date map data
Solution Approach 1:
The patent implements a feedback mechanism where vehicle-mounted sensors continuously monitor the actual surrounding environment and compare it with the map data obtained through communication. This feedback loop enables real-time validation and correction of path information, ensuring that the autonomous driving path remains accurate even when map data is not fully up-to-date.
Solution Approach 2:
The patent uses communication systems to acquire map data and path information in advance before the vehicle reaches the relevant area. This preliminary acquisition of information allows the system to prepare potential paths and make informed decisions, while the sensor data provides real-time verification.
3Speed
If sensor data is used for collision avoidance, then real-time detection is possible, but the accuracy is limited by systematic sensor constraints
Solution Approach 1:
The patent merges real-time sensor detection data with pre-acquired map data to enhance the accuracy of surrounding object detection. By combining the speed advantage of sensor-based real-time detection with the contextual accuracy of map information, the system achieves both rapid response and high precision in collision avoidance.
Data Source
AI summary
An autonomous driving device may include: a sensor configured to sense a surrounding object of an own vehicle, a controller configured to generate an autonomous driving path of the own vehicle based on movement data of the surrounding object sensed by the sensor unit, and a surrounding object analyzer configured to receive the movement data of the surrounding object from the controller and stochastically analyze an expected movement trajectory of the surrounding object. The controller may generate the autonomous driving path based on the expected movement trajectory of the surrounding object that is stochastically analyzed by the surrounding object analyzing unit.