IMAC-AgriVLN: Can Agricultural Vision-and-Language Navigation Agents be Aware of Instruction Mistakes?

2026-06-01

AI Summary

This work addresses a critical limitation in existing agricultural vision-and-language navigation (VLN) methods, which assume human instructions are always correct and thus struggle in real-world scenarios where instructions may contain errors. To enhance robustness, the authors propose the Instruction-Modality Alignment Check (IMAC) module, which evaluates the consistency between visual observations and language instructions to automatically detect and correct erroneous commands. The study introduces A2A-MI, the first agricultural VLN benchmark dataset, along with a semi-automatic annotator designed to generate diverse instruction errors. Experimental results demonstrate that integrating IMAC significantly improves task success rates under erroneous instructions, effectively narrowing the performance gap between error-prone and error-free instruction settings.

Abstract

Agricultural robots are serving as powerful assistants across a wide range of agricultural tasks, nevertheless, still heavily relying on manual operations or railway systems for movement. The AgriVLN method and the A2A benchmark pioneeringly extended Vision-and-Language Navigation (VLN) to the agricultural domain, enabling a robot to navigate to a target position following a natural language instruction. However, almost all the prior methods adopt an ideal assumption that the given instructions themselves are correct, which does not align with the realistic scenarios, because anybody may say an instruction with mistakes. To bridge this gap, we propose the A2A-MI benchmark, in which we build a semi-automatic data annotator to insert three mistake classifications into each original instruction in a more diversified and efficient way. We test several state-of-the-art agricultural VLN agents on it and observe a sufficient drop with -57% on SR and -9% on NE, from which we suggest that an agricultural VLN agent tends to assume that the given instruction is correct, so does not have the awareness to doubt it when the scenes it sees do not align with the instruction it receives. To build the awareness on instruction mistake, we propose the IMAC module analyzing the instruction and the current front-facing image, to judge whether the instruction has mistakes and attempt to correct it when needed. We integrate IMAC into the baseline model, and observe a noteworthy improvement, sufficiently narrowing the gap to the performance on instructions without mistakes. Project: https://github.com/AlexTraveling/IMAC-AgriVLN.

Problem

• Agricultural Vision-and-Language Navigation
• Instruction Mistakes
• Navigation Robustness
• AgriVLN
• Natural Language Instruction

Innovation

• instruction mistake awareness
• agricultural vision-and-language navigation
• IMAC module
• A2A-MI benchmark
• error-robust navigation