Trajectory-Based Sensor Planning for Autonomous Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomously operated vehicles, such as optionally-piloted vehicles and unmanned aerial vehicles, face challenges in navigating obstacle-rich environments due to limited peripheral field of view from directional sensors, which can hinder sharp turns and increase the need for additional, weighty and costly sensors.
Innovation Solution
Implementing trajectory-based sensor planning that dynamically adjusts the current field of view of directional sensors based on anticipated vehicle paths, using a processing subsystem to determine and initiate adjustments before changes in trajectory, allowing for optimized observation of the planned path without adding more sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If additional sensors are added to enhance peripheral field of view, then the ability to navigate autonomously in obstacle-rich environments improves, but vehicle weight and cost increase
Solution Approach 1:
The patent applies dynamics by making the sensor field of view adjustable and reconfigurable in real-time. The sensor can dynamically change its field of view width based on the vehicle's trajectory and environmental context, transitioning between narrow and wide viewing modes as needed for different navigation scenarios.
Solution Approach 2:
The system performs preliminary action by pre-planning the sensor's field of view adjustments based on the anticipated trajectory. The sensor planning module predicts future vehicle positions and orientations, and proactively configures the sensor's field of view accordingly before the vehicle reaches critical decision points.
2Adaptability or versatility
If additional sensors are added to enhance peripheral field of view, then the ability to navigate autonomously in obstacle-rich environments improves, but vehicle cost increases
Solution Approach 1:
The patent implements universality by designing a single sensor system that can perform multiple functions through dynamic reconfiguration. The same sensor provides both narrow-field detailed observation and wide-field peripheral monitoring capabilities by adjusting its field of view, eliminating the need for separate specialized sensors.
Solution Approach 2:
The system uses dynamics to enable one sensor to replace multiple static sensors through real-time field of view adjustment, reducing overall system complexity while maintaining versatile perception capabilities.
3Measurement precision
If a straight-ahead view is used to span large distances, then long-range detection is improved, but peripheral perception becomes poor
Solution Approach 1:
The sensor dynamically adjusts its field of view width based on the vehicle's operational context. During straight-line navigation, it uses a narrow field of view for long-range precision detection. When approaching turns or obstacles, it widens the field of view to capture peripheral information, seamlessly transitioning between modes.
Solution Approach 2:
The system changes the field of view parameter dynamically based on trajectory and environmental conditions. The field of view width is adjusted as a controllable parameter to optimize the balance between long-range detection precision and peripheral awareness according to the current navigation phase.
Data Source
AI summary
According to an aspect of the invention, a method of trajectory-based sensor planning for a vehicle includes receiving an indication of a planned change in a trajectory of the vehicle. A processing subsystem determines a current field of view of a directional sensor and a planned adjustment in the current field of view of the directional sensor relative to the vehicle to align with the planned change in the trajectory of the vehicle. The planned adjustment in the current field of view of the directional sensor is initiated prior to changing the trajectory of the vehicle.