Autonomous Vehicle Drifting Error Segmentation and Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving vehicles face challenges in accurately detecting and correcting drifting errors, which affect motion planning and control, especially with low-cost equipment and without high-definition maps or localization, leading to inaccuracies in path planning and control.
Innovation Solution
The system segments drifting errors into two parts using a predetermined algorithm, performing a first correction during the planning stage by modifying the trajectory's starting point and a second correction during the control stage by adjusting the heading direction, utilizing sensor data from various sensors to compensate for drifting errors and maintain a stable path.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If drifting error correction is performed using low-cost equipment without high-definition maps or localization, then the cost of the system is reduced, but the measurement precision of drifting error detection deteriorates
Solution Approach 1:
The patent introduces an intermediary computational model that predicts expected sensor readings based on the motion plan and vehicle dynamics. This predicted reading serves as a reference to compare against actual sensor data, enabling drifting error detection without requiring expensive high-definition maps or specialized localization equipment. The intermediary model bridges the gap between low-cost sensors and accurate drifting detection.
Solution Approach 2:
The patent replaces the traditional mechanical/optical localization systems (high-definition maps, specialized sensors) with a computational approach using standard sensor fusion and predictive modeling. By substituting physical infrastructure dependencies with algorithmic predictions based on vehicle dynamics and sensor data, the system achieves comparable or superior precision without the associated costs.
2Measurement precision
If drifting error detection and correction is implemented, then the navigation accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent makes existing multi-functional components perform additional functions. The sensor fusion system, already used for localization and mapping, is extended to also detect drifting errors. The motion planning system, which generates trajectories, is also used to predict expected sensor readings for drifting detection. This multi-functionality avoids adding dedicated hardware while improving navigation accuracy.
Solution Approach 2:
The system uses its own operational data (motion plans, sensor readings, vehicle state) to self-diagnose drifting errors without requiring external reference systems. The predictive model generates expected readings based on the system's own planned actions, and the comparison with actual readings enables self-detection of deviations. This self-service approach avoids external infrastructure and reduces system complexity.
3Device complexity
If drifting correction is performed only at the control stage, then the system complexity is reduced, but the manufacturing precision of path following deteriorates
Solution Approach 1:
The patent performs preliminary drifting correction by modifying the motion plan before execution. The system detects drifting errors and adjusts the trajectory in advance, so that the vehicle follows a corrected path rather than attempting to compensate during control. This preliminary action at the planning stage, combined with simple control-stage monitoring, achieves high path following accuracy without complex real-time control adjustments.
Solution Approach 2:
The patent implements a feedback loop where actual sensor readings are continuously compared with predicted readings, and the detected drifting error feeds back into both the motion planning and control systems. This feedback enables the system to learn from deviations and continuously improve path following accuracy. The feedback mechanism is simple computationally, requiring only comparison and error calculation, yet produces significant improvements in precision.
Data Source
AI summary
In one embodiment, a lateral drifting error is determined based on at least a current location of an ADV. The lateral drifting error is segmented into a first drifting error and a second drifting error using a predetermined segmentation algorithm. A planning module plans a path or trajectory for a current driving cycle (e.g., planning cycle) to drive the ADV from the current location for a predetermined period of time. The planning module performs a first drifting error correction on the trajectory by modifying at least a starting point of the trajectory based on the first drifting error to generate a modified trajectory. A control module controls the ADV to drive according to the modified trajectory, including performing a second drifting error correction based on the second drifting error.