Autonomous Vehicle Point-Turn Control in Crowded Obstacles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned ground vehicles (UGVs) face challenges in executing safe point-turns in densely crowded areas with limited maneuvering space, especially when characterized by an asymmetric contour, as they need to avoid collisions with nearby obstacles during navigation.
Innovation Solution
A computerized method and control system that uses a scanning device to generate real-time scanning output data, processing unit to create a range map, and determine conditions for a collision-free point-turn by analyzing differences between real-time range values and vehicle-contour range values, allowing the vehicle to execute a point-turn quickly without stopping.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a UGV performs a point-turn in a densely crowded area, then the vehicle can make a tight rotation in confined space, but the risk of collision with nearby obstacles increases
Solution Approach 1:
The system performs preliminary actions by calculating the vehicle contour and predicting the occupied area during point-turn in advance. The processor determines the vehicle contour based on rotation point and direction, then checks for obstacles in this predicted area before executing the turn, preventing collisions rather than reacting to them
Solution Approach 2:
The patent introduces an intermediary computational layer between the turning command and execution. The processor acts as a mediator that receives the point-turn command, calculates the vehicle contour, detects obstacles in the occupied area, and only permits execution if the area is clear, thus mediating between the turning desire and safety constraints
2Adaptability or versatility
If a UGV with asymmetric contour performs point-turn, then the vehicle can navigate through tight spaces, but the determination of safe rotation becomes more complex
Solution Approach 1:
The system applies local quality by determining the vehicle contour locally relative to each potential rotation point. Instead of using a fixed global coordinate system, the contour calculation is performed locally around the rotation point, allowing the asymmetric vehicle shape to be accurately represented in the context of each specific turning maneuver
Solution Approach 2:
The patent implements dynamics by making the vehicle contour determination dynamic and adaptive to the rotation point and direction. The contour changes based on the chosen rotation parameters, allowing the system to adapt to different turning scenarios and asymmetric vehicle configurations rather than using a static model
3Productivity
If a UGV uses traditional obstacle avoidance methods, then the vehicle can navigate in open areas, but the accuracy is insufficient when maneuvering very close to obstacles
Solution Approach 1:
The patent transitions from traditional two-dimensional sensor data to a three-dimensional vehicle contour model. By calculating the occupied area in 3D space based on rotation point, direction, and vehicle geometry, the system adds a dimensional aspect that enables accurate collision prediction even when maneuvering extremely close to obstacles
Data Source
AI summary
The presently disclosed subject matter includes a computerized method and a control system mountable on a vehicle for autonomously controlling the vehicle and enabling to execute a point-turn while avoiding collision with nearby obstacles. More specifically, the proposed technique enables an autonomous vehicle, characterized by an asymmetric contour, to execute a collision free point-turn, in an area crowded with obstacles. The disclosed method enables execution of a point turn quickly, without requiring the vehicle to stop.