Autonomous Vehicle Spin Control for Sharp Turns and Obstacles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles struggle to autonomously navigate sharp turns without operator intervention, particularly when obstacles necessitate a dynamic change in path, such as spinning in place to avoid an obstacle.
Innovation Solution
A spin controller and model predictive controller (MPC) enable an autonomous vehicle to determine the need for a spin in place and calculate an optimal spinning path to reach a destination, considering vehicle orientation and obstacles, allowing the vehicle to maneuver within 5° of the destination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If an autonomous vehicle relies on map annotations for sharp turns, then the vehicle can navigate pre-mapped turns, but the vehicle cannot dynamically respond to obstacles requiring spin maneuvers
Solution Approach 1:
The navigation system transitions from static map annotations to dynamic real-time path planning. The spin controller continuously monitors obstacle positions and dynamically generates spin maneuvers based on current vehicle state and destination, enabling adaptive response to changing environmental conditions rather than relying on pre-defined paths
Solution Approach 2:
The system uses the vehicle's own sensor data and state information to autonomously determine when and how to spin. The spin controller calculates spin parameters (angle, duration, speed) based on the vehicle's current orientation, position, and the location of obstacles, allowing the vehicle to self-navigate without external operator intervention or reliance on pre-mapped turn data
2Measurement precision
If an autonomous vehicle performs a spin in place to avoid obstacles, then the vehicle can reach the destination, but the vehicle must achieve precise angular alignment within 5° of the destination
Solution Approach 1:
The spin controller implements closed-loop feedback control by continuously monitoring the vehicle's orientation during and after the spin maneuver. The controller compares the actual final angle against the target angle (within 5° of destination) and adjusts subsequent control commands to achieve precise alignment, ensuring accurate positioning despite variations in spin execution
Solution Approach 2:
The system dynamically adjusts spin parameters (angular velocity, spin duration, steering angle) based on real-time calculations of the vehicle's state and destination requirements. By varying these parameters, the controller can achieve the precise 5° alignment tolerance while accounting for factors like current orientation, speed, and distance to destination
3Productivity
If an autonomous vehicle uses a model predictive controller to calculate spin paths, then the vehicle can optimize the spinning path, but the calculation and execution time increases
Solution Approach 1:
The model predictive controller performs partial optimization by focusing calculations on the immediate spin maneuver and short-term path adjustment rather than optimizing the entire route. This selective approach provides sufficient path optimization for the critical spin operation while minimizing computational time and avoiding excessive processing delays
Data Source
AI summary
Spin controller that can enable an autonomous vehicle (AV) to spin in place. A model predictive controller (MPC) can trigger the spin controller any time the MPC determines that a spin in place is required. In some configurations, the spin controller can move the AV to within 5° of the destination point. Spin controller can determine spin method based on the configuration of the AV, calculate an optimum turning path based at least on device mode and obstacles, and can enable the AV to spin in place.