Autonomous Vehicle Control With Multi-Controller Command Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle navigation systems face challenges in accurately determining vehicle control commands in complex environments with multiple dynamic and static objects and conditions, leading to abrupt acceleration or deceleration, affecting ride comfort and safety.
Innovation Solution
The system employs a computing device that considers various objects and conditions to generate multiple control requests, using different controllers with gain-scheduling based on distance, time, and additional information to determine weighted average accelerations for smooth navigation, incorporating sensor data and perception systems for real-time object detection and prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the vehicle uses a single control command to navigate through complex environments, then the control system is simple, but the accuracy of determining safe speeds and smooth navigation deteriorates
Solution Approach 1:
The patent divides the control system into multiple independent controllers (first controller, second controller, third controller) that each handle specific navigation tasks. Each controller generates separate control commands based on different parameters, and these commands are combined through a combiner to produce the final control output. This segmentation allows each controller to specialize in specific aspects of navigation, improving overall accuracy without requiring a single overly complex controller.
Solution Approach 2:
The patent merges multiple control commands from different controllers into a single integrated control output using a combiner. The combiner receives control commands from the first controller (based on lead object), second controller (based on speed limit), and third controller (based on closed course conditions), then combines these commands to generate the final control signal. This merging approach integrates multiple sources of control information to improve navigation accuracy while managing system complexity.
2Reliability
If the vehicle uses multiple controllers with different control commands, then the accuracy of navigation and safety improves, but the device complexity increases
Solution Approach 1:
The control system is segmented into specialized controllers, each responsible for specific navigation aspects: the first controller handles lead object following, the second controller manages speed limit compliance, and the third controller addresses closed course conditions. This segmentation improves reliability by ensuring each aspect of navigation is handled by a dedicated controller, while the modular structure manages complexity through clear division of responsibilities.
Solution Approach 2:
The combiner component serves as a universal element that integrates control commands from multiple different controllers. It handles various types of control inputs (distance-based commands, speed-based commands, course-based commands) and produces a unified control output. This multi-functional combiner reduces overall system complexity by providing a single integration point for diverse control sources.
3Measurement precision
If the vehicle determines control commands based on multiple objects and conditions, then the accuracy of speed determination improves, but the computational complexity and energy consumption increase
Solution Approach 1:
The computational workload is segmented across multiple controllers, each processing specific aspects of navigation data. The first controller processes lead object distance and speed data, the second controller processes speed limit data, and the third controller processes closed course data. This segmentation distributes computational energy consumption across multiple specialized processing units rather than requiring one controller to handle all computations, improving efficiency while maintaining accuracy.
Solution Approach 2:
Each controller performs preliminary processing of specific control aspects independently before their commands are combined. The controllers pre-process their respective inputs (distance measurements, speed measurements, course information) and generate preliminary control commands that are then integrated by the combiner. This preliminary action by specialized controllers reduces the overall computational burden by dividing the processing workload in advance.
Data Source
AI summary
Command determination for controlling a vehicle, such as an autonomous vehicle, is described. In an example, individual requests for controlling the vehicle relative to each of multiple objects or conditions in an environment are received (substantially simultaneously) and based on the request type and/or additional information associated with a request, command controllers can determine control commands (e.g., different accelerations, steering angles, steering rates, and the like) associated with each of the one or more requests. The command controllers may have different controller gains (which may be based on functions of distance, distance ratios, time to estimated collisions, etc.) for determining the controls and a control command may be determined based on the all such determined controls.