Autonomous Vehicle Coverage Path Planning for Unknown Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing coverage path planning algorithms for autonomous vehicles in unknown environments face challenges such as generating efficient paths that avoid local extrema, reducing the number of turns, and overcoming complex obstacle geometries, often resulting in longer coverage times and increased resource consumption.
Innovation Solution
The ε-STAR algorithm employs an Exploratory Turing Machine (ETM) with Multiscale Adaptive Potential Surfaces (MAPS) to guide autonomous vehicles, using sensor feedback for real-time navigation and operational commands, which generates back-and-forth coverage paths that adapt to environmental conditions and avoid local extrema by switching between multiple levels of potential surfaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional coverage path planning algorithms are used in unknown environments, then the autonomous vehicle can navigate and cover the environment, but the path length and number of turns increase, resulting in longer coverage time and higher resource consumption
Solution Approach 1:
The patent implements dynamic path planning by continuously updating the coverage path based on real-time sensor feedback and environmental changes. The algorithm dynamically adjusts the vehicle's trajectory, speed, and turning points during execution, allowing adaptive response to obstacles and changing conditions, thereby reducing unnecessary travel time and improving coverage efficiency.
Solution Approach 2:
The system employs sensor-based feedback mechanisms where onboard sensors continuously detect obstacles, map the environment, and provide real-time information to the path planning algorithm. This feedback loop enables the algorithm to optimize the coverage path dynamically, avoiding local extrema and reducing redundant movements, which directly decreases coverage time and improves productivity.
2Productivity
If traditional coverage path planning algorithms are used, then the autonomous vehicle can complete coverage, but the number of turns and trajectory length increase, leading to increased resource consumption
Solution Approach 1:
The dynamic nature of the algorithm allows it to optimize energy consumption by adjusting the vehicle's motion in real-time. By continuously recalculating the optimal path based on current environmental conditions, the system minimizes unnecessary turns and trajectory length, directly reducing the energy required for completion.
Solution Approach 2:
The algorithm changes key motion parameters such as speed, acceleration, and turning radius dynamically during path execution. By optimizing these parameters based on real-time feedback and environmental mapping, the system reduces overall energy consumption while maintaining effective coverage of the environment.
3Adaptability or versatility
If sensor-based methods are used for online planning, then the autonomous vehicle can adapt to unknown environments, but the computational complexity and device complexity increase
Solution Approach 1:
The patent segments the environment into discrete cells or grid elements, allowing the algorithm to process spatial information in manageable units. This segmentation simplifies the computational complexity by breaking down the continuous environment into discrete states, making real-time processing more feasible while maintaining adaptability to unknown environments.
Solution Approach 2:
The system performs self-mapping and self-localization using onboard sensors, eliminating the need for external infrastructure or pre-mapped environments. The algorithm autonomously builds the environmental model and plans paths based on this self-generated information, reducing the need for complex external systems while enhancing environmental adaptability.
4Loss of time
If the algorithm avoids local extrema and reduces turns, then coverage time decreases, but the path planning complexity and computational requirements increase
Solution Approach 1:
The algorithm dynamically identifies and avoids local extrema by continuously evaluating the potential for improvement in the coverage path. Through real-time assessment of environmental feedback and current trajectory, the system adapts to eliminate redundant loops and turns, reducing coverage time while managing computational complexity through efficient dynamic decision-making.
Solution Approach 2:
The algorithm maintains continuous progress toward coverage completion by ensuring that each path adjustment contributes meaningfully to reducing total coverage time. By continuously optimizing the trajectory and avoiding actions that would increase turns or trajectory length, the system achieves time reduction without excessive computational overhead.
Data Source
AI summary
Methods and apparatus related to autonomous vehicles (AVs) are provided. A mapping can be determined that tiles an environment having an AV using a plurality of cells; each cell having an environmental status. While the AV is in the environment: status data can be received relating to a location of the AV and obstacles at that location; environmental status for a cell can be updated based on the status data; a value for each cell can be determined based on the cell's environmental status; a waypoint of a coverage path that covers a region in the environment and is based on the AV's location can be determined; a determination whether the waypoint is reachable from the AV's location can be made; after determining that the waypoint is reachable, a command based on the mapping can be sent directing the AV toward the waypoint; and the waypoint can be updated.