Semi-Automated Vehicle Control System for Precise Navigation
Solution Overview
Problem
Existing unmanned vehicle control systems face challenges in precisely navigating to specific locations, such as rooms or floors within buildings, and require pre-known landing markers, which can lead to inefficiencies and vulnerabilities, especially when obstacles are small or difficult to detect.
Innovation Solution
A semi-automated control system that includes a processing circuitry and memory to establish and switch between control modes based on predetermined events, allowing for real-time control instructions from a node to navigate vehicles accurately and safely, even without pre-known markers, by determining mode changes based on geographical location and sensor inputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If automated navigation with on-board computations is used, then navigation autonomy is improved, but hardware cost and computational complexity increase
Solution Approach 1:
The patent introduces a ground-based server as an intermediary computational resource that assists the UV in navigation. The server receives sensor data from the UV, performs complex computations for path planning and obstacle avoidance, and sends control instructions back to the UV. This mediator approach allows the UV to maintain navigation autonomy while offloading heavy computational tasks to external infrastructure, thereby reducing on-board hardware requirements and costs.
2Measurement precision
If precise navigation to specific sub-locations is implemented, then delivery accuracy is improved, but control complexity increases
Solution Approach 1:
The patent divides the navigation task into multiple segments: coarse navigation to the general target area using GPS and pre-planned routes, then fine navigation to the specific sub-location (room, floor, or coordinate) using real-time sensor data and iterative optimization. This segmentation allows the system to achieve high delivery accuracy by breaking down the complex control problem into manageable stages, reducing overall control complexity.
3Reliability
If automated obstacle detection systems are used, then navigation safety is improved, but detection capability deteriorates for small or difficult-to-detect obstacles
Solution Approach 1:
The patent combines multiple sensor types (cameras, LIDAR, ultrasonic sensors, infrared sensors) into an integrated obstacle detection system. Each sensor type detects different characteristics of obstacles, and their data is fused to create a comprehensive view of the environment. This merging of detection modalities significantly improves the system's ability to detect small or difficult-to-detect obstacles while maintaining navigation safety through redundant verification.
Data Source
AI summary
An apparatus and method for semi-automated control of a vehicle. The apparatus includes: a processing circuitry; and a memory, the memory containing instructions that, when executed by the processing circuitry, configure the apparatus to: establish control of the vehicle, wherein establishing the control further comprises determining a set of instructions for controlling the vehicle, wherein the apparatus is configured to control the vehicle based on the set of instructions during a first mode of control; receive at least one control instruction from a node, wherein the apparatus is configured to control, in real-time, the vehicle based on the set of instructions and the at least one control instruction received from the node during a second mode of control; determine, based on a predetermined list of mode changing events, a mode changing event; and change from the first mode to the second mode, when the mode changing event is determined.