Autonomous Rideshare Local Assistance for Vehicle Immobilization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles may become immobilized or unable to make decisions due to connectivity issues or operational difficulties, leading to delayed remote assistance and undesirable situations, as they struggle with classifying objects or navigating obstacles.
Innovation Solution
An autonomous vehicle local assistance system allows passengers to provide direct input to the vehicle's decision-making process by presenting them with options that have confidence values exceeding a threshold, enabling them to assist in resolving immobility or perception issues, while preventing malicious interference and incorporating passenger trust scores to adjust the level of input availability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If remote assistance is used to resolve autonomous vehicle immobilization, then the vehicle can receive expert help, but the response time is delayed due to connectivity issues and operator availability limits
Solution Approach 1:
The system introduces a local assistance intermediary (the passenger) who acts as a mediator between the autonomous vehicle system and the decision-making process. When the vehicle encounters difficult situations, the local assistance module presents perceived objects and potential actions to the passenger, who can then provide input to resolve the immobilization situation without waiting for remote operator intervention.
Solution Approach 2:
The assistance system is segmented into multiple levels: autonomous decision-making for routine situations, local assistance for difficult situations, and remote assistance for complex cases. This segmentation allows the vehicle to quickly handle most situations autonomously while providing structured escalation paths for more challenging scenarios, reducing overall response time.
2Productivity
If passenger input is allowed to resolve autonomous vehicle decisions, then the system can handle difficult situations more quickly, but there is risk of malicious or erroneous interference
Solution Approach 1:
The system implements feedback mechanisms where passenger input is continuously evaluated against the autonomous system's perceptions and decisions. The local assistance module provides feedback loops that allow the system to learn from passenger corrections while maintaining safety checks, ensuring that passenger input improves rather than degrades system performance.
Solution Approach 2:
The system dynamically adjusts the level of passenger involvement based on the situation's complexity and the passenger's trust score. For routine situations, the system operates autonomously without passenger input. For difficult situations, the system selectively engages the passenger only when needed, and the degree of passenger control varies based on their established trust level with the system.
3Adaptability or versatility
If the autonomous system presents all possible options to the passenger, then the passenger can provide comprehensive input, but the interface becomes complex and overwhelming
Solution Approach 1:
The interface presents information with varying levels of detail and interactivity based on the local context of each situation. For high-confidence perceptions, the system presents simple confirmation options. For low-confidence or difficult situations, the system provides more detailed information and multiple action options, adapting the interface complexity to match the situation's complexity rather than presenting all possible options uniformly.
Solution Approach 2:
The system selectively presents only the most relevant options to the passenger based on the current situation's needs and the autonomous system's confidence levels. Rather than presenting all possible actions, the system filters and prioritizes options that are most likely to be useful, providing sufficient choice without overwhelming the passenger with unnecessary alternatives.
Data Source
AI summary
A method is described and includes subsequent to an autonomous vehicle becoming immobilized, initiating a local assistance request; subsequent to the initiating, receiving local assistance input from a passenger of the autonomous vehicle; and using the local assistance input to determine an action to be taken by the autonomous vehicle to mobilize the autonomous vehicle.