Autonomous Vehicle Spin Control for Sharp Turns and Obstacle Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in navigating sharp turns without human intervention, relying on pre-mapped annotations and lacking the ability to dynamically avoid obstacles by spinning in place to find an alternate path.
Innovation Solution
A system and method that includes a spin controller, triggered by a model predictive controller, allowing the autonomous vehicle to spin in place by determining the axis of rotation and calculating angles for clockwise or counterclockwise rotation, and selecting a path based on the smallest total cost, which can be adjusted to avoid obstacles and ensure safe navigation through sharp turns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the autonomous vehicle relies on pre-mapped annotations for sharp turns, then the navigation can be planned in advance, but the vehicle cannot dynamically avoid obstacles by spinning in place
Solution Approach 1:
The navigation system transitions from static pre-mapped annotations to dynamic real-time path planning. The model predictive controller continuously calculates optimal paths based on current vehicle state and detected obstacles, enabling the vehicle to adapt to changing environments by spinning in place when obstacles are detected, rather than relying on predetermined turn annotations.
Solution Approach 2:
The vehicle equips itself with onboard sensors and computing capabilities to independently detect obstacles and calculate alternative paths in real-time. The spin controller and model predictive controller work together to autonomously determine when and how to spin the vehicle to avoid obstacles without requiring external intervention or pre-programmed solutions for every possible scenario.
2Reliability
If the autonomous vehicle uses traditional turning methods, then the system is simpler to implement, but the vehicle cannot navigate sharp turns or avoid obstacles effectively
Solution Approach 1:
The control system is divided into specialized modules: a model predictive controller for high-level path planning, a spin controller for executing in-place rotation maneuvers, and coordinate transformation modules for converting between different reference frames. This segmentation allows each component to focus on a specific aspect of navigation, improving reliability for complex maneuvers while managing overall system complexity through modular design.
Solution Approach 2:
The spin controller acts as an intermediary between the model predictive controller's path planning and the vehicle's physical execution. It translates desired path changes into specific spin maneuvers, coordinating the vehicle's rotation and translation to achieve precise navigation through sharp turns and around obstacles that would be impossible with traditional turning methods.
3Adaptability or versatility
If the autonomous vehicle spins in place to avoid obstacles, then the vehicle can find alternate paths dynamically, but the control complexity increases significantly
Solution Approach 1:
The model predictive controller continuously monitors the vehicle's position, orientation, and detected obstacles, using this feedback to recalculate optimal paths in real-time. When obstacles are detected, the system feedback-drivenly determines the appropriate spin magnitude and direction, then verifies the maneuver's effectiveness by continuing to monitor the vehicle's state and obstacle positions, adjusting the path as needed.
Solution Approach 2:
The control system dynamically changes key parameters during spin maneuvers, including the vehicle's angular velocity, orientation angle, and linear velocity. The model predictive controller adjusts these parameters based on the calculated optimal path, enabling the vehicle to execute precise spin maneuvers of varying magnitudes and directions to navigate around obstacles and reach alternative destinations.
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.