Automated Driving Path Planning With Off-Board Augmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The limitations of existing Automated Driving Systems (ADS) due to hardware and power resource constraints restrict the processing of raw sensor data and sophisticated algorithms, limiting the addition of new functionalities without increasing size, power consumption, or cost.
Innovation Solution
A method that involves locally processing sensor data to generate a local world-view and a candidate path, transmitting data to a remote system for supplementary processing, and selecting a path based on constraints, allowing for advanced algorithms and models to be executed off-board, thereby augmenting ADS capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If more sophisticated algorithms and more input data are used for perception and path planning, then the quality and safety of automated driving decisions are improved, but the processing capability requirements increase, leading to larger size, higher power consumption, and increased cost of on-board hardware
Solution Approach 1:
The path planning function is segmented into two parts: local candidate path generation performed on-board by the vehicle's perception module, and supplementary candidate path generation performed remotely by a server. This segmentation allows the on-board system to use simpler, less resource-intensive algorithms while still achieving high-quality path planning through the combination of local and remote processing results.
Solution Approach 2:
A communication interface acts as an intermediary between the on-board vehicle system and the remote server. The interface transmits necessary data to the server and receives supplementary path information, enabling the vehicle to access enhanced processing capabilities without permanently hosting the complex processing infrastructure on-board.
2Loss of information
If more input data from sensors is processed, then the awareness of vehicle surroundings and the quality of automated decisions are improved, but the amount of data that can be effectively utilized is limited by hardware and power resources
Solution Approach 1:
The invention extracts the computationally intensive data processing tasks from the on-board vehicle system and relocates them to a remote server. The vehicle's perception module collects sensor data and transmits it to the server, which performs the sophisticated processing and returns results. This extraction allows the vehicle to utilize more sensor data for better environmental awareness without proportionally increasing on-board power consumption.
3Adaptability or versatility
If the on-board system is already at its capability limit, then adding new functionality or extensions is restricted, but there is a need to improve performance and extend functionality
Solution Approach 1:
The invention adds a new dimension to the path planning system by introducing remote server processing in addition to on-board processing. This dimensional expansion allows the system to access enhanced computational capabilities and sophisticated algorithms without modifying or upgrading the on-board hardware, thereby extending functionality while maintaining the existing on-board system capability limits.
Data Source
AI summary
The present disclosure relates to a method for augmenting capabilities of an Automated Driving System (ADS) of a vehicle. The method includes locally processing, by means of a perception module of the ADS, sensor data obtained from one or more sensors of the vehicle in order to generate a local world-view of the ADS. The sensor data is associated with a time period and includes information about a surrounding environment of the vehicle during the time period. The method further includes generating a local candidate path to be executed by the ADS based on the generated local world-view of the ADS, and transmitting a first set of data to a remote system. The first set of data is associated with the time period and including information about the surrounding environment of the vehicle during the time period.