Infrastructure-Based Tactical Planning for Autonomous Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous vehicle control systems are not sufficiently safe in complex traffic situations, particularly in inner-city intersections, due to limited tactical decision-making capabilities and reliance on onboard sensors, which restrict their ability to adapt to diverse and dynamic environments.
Innovation Solution
A system that utilizes a static detector device to generate object information from the environment, combined with a static planning device for determining driving tactics, allowing for anticipatory and location-specific tactical planning by incorporating information from infrastructure-based detectors and machine learning to select suitable driving strategies based on safety, time, and environmental criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If tactical planning is performed solely by the vehicle using onboard sensors and learned behaviors, then the system operates independently without infrastructure support, but the tactical decision-making capability is limited and insufficient for complex traffic situations
Solution Approach 1:
The patent merges the vehicle's onboard tactical planning system with an infrastructure-based tactical planning system. The infrastructure component uses static detectors and machine learning models to generate driving tactics that are transmitted to the vehicle, combining external environmental understanding with the vehicle's own sensor data and decision-making capabilities.
Solution Approach 2:
The infrastructure-based tactical planning system acts as an intermediary between the environment and the vehicle's control system. It processes environmental data from static detectors and generates driving tactics that mediate between raw sensor information and the vehicle's execution systems, enhancing the vehicle's decision-making in complex situations.
2Reliability
If infrastructure-based detectors are used to extend the field of view and provide environmental information, then traffic safety and flow are improved, but the limited tactical decision-making possibilities of planning algorithms in the vehicle remain unsolved
Solution Approach 1:
The patent extracts the tactical planning function from the vehicle and relocates it to the infrastructure. The infrastructure-based system uses static detectors and machine learning to determine driving tactics, which are then transmitted to the vehicle. This extraction reduces the computational burden and decision-making limitations of the vehicle's onboard systems while maintaining safety improvements.
3Ease of operation
If the vehicle uses learned tactical behaviors from individual driving maneuvers, then it can make tactical decisions based on available information, but it is restricted to the ego perspective and cannot anticipate situations beyond sensor range
Solution Approach 1:
The infrastructure-based tactical planning system performs preliminary analysis of the environment using static detectors and machine learning models before the vehicle arrives. It generates driving tactics in advance that incorporate environmental information beyond the vehicle's sensor range, allowing the vehicle to receive pre-computed tactical guidance that anticipates future situations.
Data Source
AI summary
A method and system for autonomously controlling an autonomous vehicle, wherein the system includes a stationary detector device for statically generating object information from an environment of the detector device and includes a stationary planning device which is configured to determine a statically generated driving tactic for the autonomous vehicle based on the object information, where the autonomous vehicle has a mobile strategy unit for generating a driving strategy of the autonomous vehicle and a mobile planning unit for determining a driving tactic based on the predefined driving strategy, a current driving situation around the autonomous vehicle, and the driving tactic determined by the stationary planning device.