System-Level Analysis of Module Uncertainty Quantification in the Autonomy Pipeline
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908786986319872 |
|---|---|
| author | Deglurkar, Sampada Shen, Haotian Muthali, Anish Pavone, Marco Margineantu, Dragos Karkus, Peter Ivanovic, Boris Tomlin, Claire J. |
| author_facet | Deglurkar, Sampada Shen, Haotian Muthali, Anish Pavone, Marco Margineantu, Dragos Karkus, Peter Ivanovic, Boris Tomlin, Claire J. |
| contents | Modern autonomous systems with machine learning components often use uncertainty quantification to help produce assurances about system operation. However, there is a lack of consensus in the community on what uncertainty is and how to perform uncertainty quantification. In this work, we propose that uncertainty measures should be understood within the context of overall system design and operation. To this end, we present two novel analysis techniques. First, we produce a probabilistic specification on a module's uncertainty measure given a system specification. Second, we propose a method to measure a system's input-output robustness in order to compare system designs and quantify the impact of making a system uncertainty-aware. In addition to this theoretical work, we present the application of these analyses on two real-world autonomous systems: an autonomous driving system and an aircraft runway incursion detection system. We show that our analyses can determine desired relationships between module uncertainty and error, provide visualizations of how well an uncertainty measure is being used by a system, produce principled comparisons between different uncertainty measures and decision-making algorithm designs, and provide insights into system vulnerabilities and tradeoffs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_12019 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | System-Level Analysis of Module Uncertainty Quantification in the Autonomy Pipeline Deglurkar, Sampada Shen, Haotian Muthali, Anish Pavone, Marco Margineantu, Dragos Karkus, Peter Ivanovic, Boris Tomlin, Claire J. Systems and Control Modern autonomous systems with machine learning components often use uncertainty quantification to help produce assurances about system operation. However, there is a lack of consensus in the community on what uncertainty is and how to perform uncertainty quantification. In this work, we propose that uncertainty measures should be understood within the context of overall system design and operation. To this end, we present two novel analysis techniques. First, we produce a probabilistic specification on a module's uncertainty measure given a system specification. Second, we propose a method to measure a system's input-output robustness in order to compare system designs and quantify the impact of making a system uncertainty-aware. In addition to this theoretical work, we present the application of these analyses on two real-world autonomous systems: an autonomous driving system and an aircraft runway incursion detection system. We show that our analyses can determine desired relationships between module uncertainty and error, provide visualizations of how well an uncertainty measure is being used by a system, produce principled comparisons between different uncertainty measures and decision-making algorithm designs, and provide insights into system vulnerabilities and tradeoffs. |
| title | System-Level Analysis of Module Uncertainty Quantification in the Autonomy Pipeline |
| topic | Systems and Control |
| url | https://arxiv.org/abs/2410.12019 |