Vehicle Trajectory Control in Degraded Visual Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to effectively control vehicles in degraded visual environments due to unreliable sensor data, which impairs obstacle identification and navigation, particularly during takeoff and landing phases, affecting both autonomous and manual operations.
Innovation Solution
A method and system that identify a degraded visual environment by comparing sensor data with expected data, determining a first trajectory segment to search for an improved navigation environment, and adjusting to a second segment based on whether the environment is improved, using LIDAR and other sensors to guide the vehicle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle operates in a degraded visual environment using conventional navigation methods, then the vehicle can maintain its route, but the navigation reliability deteriorates due to unreliable sensor data and impaired obstacle identification
Solution Approach 1:
The trajectory is divided into multiple segments (first segment and second segment) based on the phase of the route. The system segments the navigation path into a search phase (first segment) where the vehicle searches for improved visual conditions, and a execution phase (second segment) where the vehicle completes the maneuver. This segmentation allows adaptive navigation strategies tailored to specific route phases, improving reliability in degraded visual environments.
Solution Approach 2:
The system dynamically adjusts the trajectory segments based on real-time sensor data quality and visual environment conditions. The controller modifies navigation behavior adaptively - searching for improved visual conditions during the first segment, then transitioning to the second segment based on whether visual improvement was achieved. This dynamic adjustment optimizes navigation reliability under varying environmental conditions.
2Reliability
If the vehicle searches for improved visual environment conditions, then navigation reliability may improve, but the time required to complete the route phase increases
Solution Approach 1:
The system performs a limited search for improved visual conditions along the first segment trajectory rather than extensively searching all possible areas. The search is partial - sufficient to identify improved conditions when available - but constrained to maintain route completion efficiency. This balanced approach achieves adequate obstacle identification reliability without excessive time loss.
Solution Approach 2:
The route is segmented into phases with different objectives: the first segment focuses on searching for improved visual conditions, while the second segment focuses on completing the navigation maneuver. This temporal and functional segmentation allows the system to allocate time appropriately - searching when necessary, then executing the primary navigation task, thereby minimizing overall time loss while maintaining reliability.
3Measurement precision
If the system uses multiple sensors to detect degraded visual environment, then detection accuracy improves, but the system complexity increases
Solution Approach 1:
The system employs multiple sensors (LIDAR, cameras, radar) that serve dual purposes: they detect degraded visual environment conditions and simultaneously provide navigation and obstacle identification data. This multi-functionality improves detection accuracy for degraded environments without proportionally increasing system complexity, as the same hardware performs multiple functions.
Solution Approach 2:
The system merges data from multiple sensor sources into a unified degraded environment detection process. Rather than treating each sensor independently, the controller integrates LIDAR, camera, and radar data to collectively identify degraded visual conditions. This merging approach improves measurement precision while managing system complexity through integrated processing.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables adaptive and reliable navigation in degraded visual environments by searching for improved conditions and adjusting trajectories, enhancing safety and operational efficiency during challenging visual conditions.
Implementation Method 1
the one or more sensors comprise a Light Detection and Ranging (LIDAR) device, and determining the difference between the sensor data and the expected data comprises determining a difference between a number of returning light pulses represented by the sensor data and an expected number of returning light pulses
Data Source
Figure 1A
Figure 1B
Figure 2A
AI summary
In an example, a method for controlling a vehicle in a degraded visual environment is provided. The method includes identifying a degraded visual environment corresponding to a phase of a route followed by the vehicle. The method includes determining, based on the phase of the route, a first segment of a trajectory of the vehicle along which to search for a location with an improved navigation environment. The method includes causing the vehicle to follow the first segment until: (i) identifying the improved navigation environment, or (ii) reaching an end of the first segment without identifying the improved navigation environment. The method includes determining a second segment of the trajectory based on whether the improved navigation environment has been identified. The method includes causing the vehicle to follow the second segment.