Research map / Large language models / Agents, code and math
Explainable Robot Navigation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
A doctoral consortium thesis proposal on explainable robot navigation that combines machine learning (affordance recognition) with AI planning, modeling explanation generation at path-planning and task-planning abstraction levels plus user studies of explanation preferences.
Problem
Autonomous mobile robots in dynamic, social environments often cannot provide explanations for their navigation actions that align with human reasoning and expectations.
Approach
A hierarchical hybrid framework: affordance-based machine learning for low-level obstacle/target explanations, combined with AI-planning-based explanation generation (explanations as plan actions) at the high task level, plus studies of explanation modality and human preferences.
Key results
- Reports that in most situations people prefer visual-textual explanations over visual-only or textual-only explanations (Halilovic, Chandrayan, and Krivic 2024).
- Reports that inherently explainable models can supply robots with knowledge of how much each obstacle contributes to the robot's state, enabling visual explanations (Halilovic and Lindner 2022).
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use visual-textual explanations instead of visual or textual explanations alone | robot navigation explanations · explanation situations | user modality preference | Helped |
Limitations
- Results are summarized from prior work; the thesis proposal reports no new experiments.
- Unified hierarchical framework is still at planned stage.
Experimental setup
Metrics: understanding satisfaction trust
Builds on: explicable planning model reconciliation
Closest papers by meaning
- Even-if Explanations: Formal Foundations, Priorities and Complexity (AAAI 2025 · similarity 0.84)
- Position: Explainable AI Cannot Advance Without Better User Studies (ICML 2025 · similarity 0.84)
- Towards Trustworthy Autonomous Systems via Conversations and Explanations (AAAI 2024 · similarity 0.83)
- Everything, Everywhere, All at Once: Is Mechanistic Interpretability Identifiable? (ICLR 2025 · similarity 0.83)
- Autonomous Policy Explanations for Effective Human-Machine Teaming (AAAI 2024 · similarity 0.83)
- BEE: Metric-Adapted Explanations via Baseline Exploration-Exploitation (AAAI 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.