Vehicle Control Training for Autonomous Navigation in Learned Areas
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle navigation systems require manual navigation to multiple destinations within an environment, which is time-consuming and inefficient, especially when reversing along a predetermined path.
Innovation Solution
A control system for vehicles that allows autonomous navigation to a second destination by utilizing vehicle control data collected during a first maneuver, identifying a navigable area with multiple possible paths, and using sensors to gather environment data for subsequent autonomous navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual navigation is performed to multiple destinations, then navigation flexibility is improved, but time consumption increases
Solution Approach 1:
The system performs preliminary actions by recording control inputs and environment data during manual navigation phases. This recorded data is then reused for autonomous navigation to multiple destinations, eliminating the need to manually navigate to each destination while preserving navigation flexibility through the stored control sequences
2Productivity
If autonomous navigation uses recorded control data from first maneuver to reach second destination, then productivity is improved, but device complexity increases
Solution Approach 1:
The control system is designed with multi-functionality to handle both manual and autonomous navigation modes. The same control system records control inputs during manual operation and then replays this recorded data for autonomous navigation, allowing a single system to perform multiple functions without requiring separate dedicated systems for each mode
3Loss of time
If the system records and replays control inputs for navigation, then loss of time is reduced, but measurement precision requirements increase
Solution Approach 1:
The system incorporates feedback mechanisms where sensors continuously monitor the current environment during autonomous navigation and compare it with the recorded environment data. This feedback allows the system to detect deviations and make necessary adjustments, ensuring accurate navigation despite the reliance on pre-recorded control inputs
Data Source
AI summary
Embodiments of the present invention provide a control system (100). The control system (100) is for a host vehicle (10) operable in an autonomous mode and a non-autonomous mode. The control system (100) comprises one or more controllers (110). The control system (100) is configured to receive, when operating in a non-autonomous mode, a mode signal and environment data. The mode signal is indicative of the host vehicle (10) operating in a training mode. The environment data is indicative of a sensed environment of the host vehicle (10) during a first manoeuvre by the host vehicle (10) from a first location to a first navigation goal. In the training mode, the one or more controllers (110) identify a navigable area and output vehicle control data. The navigable area is in a vicinity of the first manoeuvre and is suitable to contain a plurality of possible navigation paths for subsequent navigation of the host vehicle (10), operating in an autonomous mode, to a second navigation goal within the navigable area. The second navigation goal is different to the first navigation goal. The navigable area is identified in dependence on the environment data. The vehicle control data is indicative of the navigable area. When subsequently navigating to the second navigation goal in the autonomous mode, the control system (100) utilises the vehicle control data to autonomously control the host vehicle (10).