<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Huan Ning on Huan Ning | Geospatial AI and Autonomous GIS</title><link>https://www.huanning.me/</link><description>Recent content in Huan Ning on Huan Ning | Geospatial AI and Autonomous GIS</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Sun, 25 Oct 2026 23:33:00 -0400</lastBuildDate><atom:link href="https://www.huanning.me/index.xml" rel="self" type="application/rss+xml"/><item><title>A Coding Agent for Computer-Vision Algorithm Development: Removing the Jello Effect from Drone Images</title><link>https://www.huanning.me/project/jello-correction/</link><pubDate>Sun, 25 Oct 2026 23:33:00 -0400</pubDate><guid>https://www.huanning.me/project/jello-correction/</guid><description>Correcting rolling-shutter &amp;ldquo;jello&amp;rdquo; distortion in drone images with an AI coding agent, and lessons on where the agent helps and where it still needs a human driver.</description></item><item><title>Research Architect: An AI Copilot for Human-AI Scientific Inquiry</title><link>https://www.huanning.me/project/research-architect/</link><pubDate>Sun, 25 Oct 2026 22:35:00 -0400</pubDate><guid>https://www.huanning.me/project/research-architect/</guid><description>An AI copilot for human-AI research collaboration that frames grounded, testable hypotheses around a shared influence graph.</description></item><item><title>Global Street Tree Diameter Growth Monitoring</title><link>https://www.huanning.me/project/street-tree-diameter/</link><pubDate>Sun, 25 Oct 2026 21:36:43 -0400</pubDate><guid>https://www.huanning.me/project/street-tree-diameter/</guid><description>Measuring street tree trunks year after year in historical street view imagery to track how fast city trees grow.</description></item><item><title>AI in the Shipping Industry: A Practical Application of Drones and Vision Language Models</title><link>https://www.huanning.me/project/ship-draft-drones/</link><pubDate>Sun, 25 Oct 2026 16:35:22 -0400</pubDate><guid>https://www.huanning.me/project/ship-draft-drones/</guid><description>Drones photograph a cargo ship&amp;rsquo;s draft marks and a vision language model reads them, cutting the inspection from more than an hour to about 15 minutes.</description></item><item><title>AI-Generated Synthetic Users for Software Testing</title><link>https://www.huanning.me/project/synthetic-users/</link><pubDate>Sun, 25 Oct 2026 16:26:12 -0400</pubDate><guid>https://www.huanning.me/project/synthetic-users/</guid><description>Fictional users, each with a simulated year of life, for testing the memory system of an AI companion without using private conversations.</description></item><item><title>How to build a memory system for an AI agent?</title><link>https://www.huanning.me/post/ai-agent-memory-system/</link><pubDate>Tue, 13 Oct 2026 20:05:24 -0500</pubDate><guid>https://www.huanning.me/post/ai-agent-memory-system/</guid><description>In my earlier work on autonomous GIS, I often wanted to create a memory system for AI agents that could build on previous interactions instead of starting over with every task. I did not have the opportunity or time to fully explore that memory system then. Recently, I have returned to this question from another […]</description></item><item><title>New writing: Memory systems for AI agents</title><link>https://www.huanning.me/news/new-article-agent-memory/</link><pubDate>Tue, 13 Oct 2026 00:00:00 +0000</pubDate><guid>https://www.huanning.me/news/new-article-agent-memory/</guid><description>A new blog essay explores checkable conversational memory and its applications to autonomous research agents.</description></item><item><title>Search</title><link>https://www.huanning.me/search/</link><pubDate>Tue, 13 Oct 2026 00:00:00 +0000</pubDate><guid>https://www.huanning.me/search/</guid><description/></item><item><title>Tutorial video for Research Architect, an AI copilot for human-AI teaming research</title><link>https://www.huanning.me/post/2026-09-14-research-architect-tutorial/</link><pubDate>Mon, 14 Sep 2026 00:00:00 +0000</pubDate><guid>https://www.huanning.me/post/2026-09-14-research-architect-tutorial/</guid><description>Last month, we released Research Architect, an AI copilot for human-AI research collaboration focused on hypothesis framing. Prof. Xiao Huang created a 3-minute video tutorial on how to use it.</description></item><item><title>What Should Be the Scientific Community’s Immediate Priorities in the Age of AGI?</title><link>https://www.huanning.me/thoughts/2026-09-03-scientific-priorities-agi/</link><pubDate>Thu, 03 Sep 2026 20:40:11 -0500</pubDate><guid>https://www.huanning.me/thoughts/2026-09-03-scientific-priorities-agi/</guid><description>On September 3, 2026, OpenAI released GPT-6 and formally declared that the era of artificial general intelligence, or AGI, had arrived, meaning that artificial intelligence now surpasses most humans in most domains. Regardless of how AGI is defined, AI has clearly reached an exceptionally high level of analytical and reasoning capability, at a cost that […]</description></item><item><title>Federal Geographic Data Committee adopted our proposed goals of Autonomous GIS</title><link>https://www.huanning.me/news/autonomous-gis-sdi/</link><pubDate>Sat, 22 Aug 2026 12:02:39 -0500</pubDate><guid>https://www.huanning.me/news/autonomous-gis-sdi/</guid><description>We recently learned that the U.S. Federal Geographic Data Committee (FGDC) adopted our proposed Autonomous GIS goals as part of its Spatial Data Infrastructure Modernization 2026-2027 objectives. SDI Artificial Intelligence (SDI-AI) – SDI architecture and standards, research and development of GEOAI, AIAgents, large language model’s (LLM’s) to support self-generating, self-orga</description></item><item><title>Federal Geographic Data Committee adopted our proposed goals of Autonomous GIS</title><link>https://www.huanning.me/post/federal-geographic-data-committee-adopted-our-proposed-goals-of-autonomous-gis/</link><pubDate>Sat, 22 Aug 2026 12:02:39 -0500</pubDate><guid>https://www.huanning.me/post/federal-geographic-data-committee-adopted-our-proposed-goals-of-autonomous-gis/</guid><description>We recently learned that the U.S. Federal Geographic Data Committee (FGDC) adopted our proposed Autonomous GIS goals as part of its Spatial Data Infrastructure Modernization 2026-2027 objectives. SDI Artificial Intelligence (SDI-AI) – SDI architecture and standards, research and development of GEOAI, AIAgents, large language model’s (LLM’s) to support self-generating, self-orga</description></item><item><title>An AI Copilot for human-AI scientific inquiry: Research Architect</title><link>https://www.huanning.me/post/2026-08-10-research-architect/</link><pubDate>Mon, 10 Aug 2026 00:00:00 +0000</pubDate><guid>https://www.huanning.me/post/2026-08-10-research-architect/</guid><description>Our team is pleased to release Research Architect, an AI copilot for human-AI research collaboration, specifically for hypothesis framing.</description></item><item><title>Can a 31B open-weight model handle Autonomous GIS tasks on a single GPU? I tested Google’s new Gemma-4-31b-it-Q4-K-M on an RTX 3090 to find out. 🤖🗺️</title><link>https://www.huanning.me/post/can-a-31b-open-weight-model-handle-autonomous-gis-tasks-on-a-single-gpu-i-tested-googles-new-gemma-4-31b-it-q4-k-m-/</link><pubDate>Mon, 06 Apr 2026 21:02:55 -0500</pubDate><guid>https://www.huanning.me/post/can-a-31b-open-weight-model-handle-autonomous-gis-tasks-on-a-single-gpu-i-tested-googles-new-gemma-4-31b-it-q4-k-m-/</guid><description>I used my Autonomous GIS agent, LLM-Geo (github.com/gladcolor/LLM-Geo), to run a school walkability case study. The pipeline task: 1️⃣ Generate 1 km study area buffers around multiple schools. 2️⃣ Compute walkability scores (the ratio of sidewalk length to road length). 3️⃣ Map the results with the score as the title, overlaying sidewalks and buffers on […]</description></item><item><title>From Data Source Handbooks to Agent Skills: How Academic Research Anticipated Industry Trends</title><link>https://www.huanning.me/thoughts/2026-02-19-agent-skills/</link><pubDate>Thu, 19 Feb 2026 21:42:28 -0500</pubDate><guid>https://www.huanning.me/thoughts/2026-02-19-agent-skills/</guid><description>In May 2024, our team (Zhenlong Li Temitope Ezekiel Akinboyewa M. Naser Lessani) developed LLM-Find — an autonomous GIS agent framework for geospatial data retrieval. At its core is a concept we call the “data source handbook”: a structured, plug-and-play knowledge package that teaches an LLM how to retrieve data from a specific source, including […]</description></item><item><title>Parquet format may be faster than CSV</title><link>https://www.huanning.me/post/2025-10-14-parquet-vs-csv/</link><pubDate>Tue, 14 Oct 2025 20:33:31 -0500</pubDate><guid>https://www.huanning.me/post/2025-10-14-parquet-vs-csv/</guid><description>I use CSV file for a long time to process a lot of tabular data. However, one of my recent project need to extract rows from a large parquet table. After a few testing, I found that DuckDB can speed up my work by about 5 times. I recommend to try Parquet and DuckDB if […]</description></item><item><title>AI-assisted scientific research</title><link>https://www.huanning.me/thoughts/2025-09-11-ai-research/</link><pubDate>Thu, 11 Sep 2025 13:01:39 -0500</pubDate><guid>https://www.huanning.me/thoughts/2025-09-11-ai-research/</guid><description>Recently, Google released a wonderful paper: An AI system to help scientists write expert-level empirical software. The authors synthesized the following ideas to enable LLM to write programs for scientific research: Genetic Programming, Generative Programming, LLM coding, Automated Machine Learning (AutoML), and agents for scientific discovery.</description></item><item><title>Autonomous GIS as infrastructure?</title><link>https://www.huanning.me/thoughts/2025-04-13-gis-infrastructure/</link><pubDate>Sun, 13 Apr 2025 15:19:52 -0500</pubDate><guid>https://www.huanning.me/thoughts/2025-04-13-gis-infrastructure/</guid><description>In a recent vision paper, GIScience in the Era of Artificial Intelligence: A Research Agenda Towards Autonomous GIS, we proposed three scales of autonomous GIS agents: local, centralized, and infrastructure. Some GIS agents have been developed on local and centralized scales, such as LLM-Geo and Google Geospatial Reasoning. What do the infrastructure agents look like?</description></item><item><title>Geospatial Reasoning by Google: A Leap Towards Autonomous GIS</title><link>https://www.huanning.me/thoughts/2025-04-10-geospatial-reasoning/</link><pubDate>Thu, 10 Apr 2025 10:02:53 -0500</pubDate><guid>https://www.huanning.me/thoughts/2025-04-10-geospatial-reasoning/</guid><description>Google Research released a framework named Geospatial Reasoning, which provides a user-friendly inference for spatial analysis and visualization. By receiving users&amp;rsquo; requests in natural language about geospatial tasks, Geospatial Reasoning can choose appropriate data sources and foundation models and then create and execute geoprocessing workflows to extract information and insights from data to reply to users&amp;rsquo; requests.</description></item><item><title>Building Large Spatial Models</title><link>https://www.huanning.me/thoughts/2025-04-01-building-large-spatial-models/</link><pubDate>Tue, 01 Apr 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/thoughts/2025-04-01-building-large-spatial-models/</guid><description>What a large spatial model could be, how geospatial embedding might represent places over time, and the computing, privacy, and data challenges of building one.</description></item><item><title>Estimating Hourly Neighborhood Population Using Mobile Phone Data in the United States</title><link>https://www.huanning.me/publication/hourly-population/</link><pubDate>Tue, 01 Apr 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/publication/hourly-population/</guid><description>Hourly neighborhood-level population estimation using nationwide human mobility data.</description></item><item><title>Hourly population research and interactive map</title><link>https://www.huanning.me/news/hourly-population-map/</link><pubDate>Tue, 01 Apr 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/news/hourly-population-map/</guid><description>A 2025 preprint presents a neighborhood-scale U.S. hourly population dataset, with an interactive map.</description></item><item><title>U.S. Hourly Population Maps</title><link>https://www.huanning.me/project/hourly-population/</link><pubDate>Tue, 01 Apr 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/project/hourly-population/</guid><description>Fine-grained estimates of hourly population in roughly 220,000 U.S. census block groups during 2022.</description></item><item><title>GIScience in the Era of Artificial Intelligence: A Research Agenda Towards Autonomous GIS</title><link>https://www.huanning.me/post/agenda-towards-autonomous-gis/</link><pubDate>Mon, 31 Mar 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/post/agenda-towards-autonomous-gis/</guid><description>A research agenda exploring Autonomous GIS, AI agents, and future geographic information infrastructure.</description></item><item><title>GIScience in the Era of Artificial Intelligence: A Research Agenda Towards Autonomous GIS</title><link>https://www.huanning.me/publication/autonomous-gis-agenda/</link><pubDate>Mon, 31 Mar 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/publication/autonomous-gis-agenda/</guid><description>A collaborative research agenda for the future of AI-enabled geographic information science.</description></item><item><title>The story behind Autonomous GIS</title><link>https://www.huanning.me/post/2025-03-29-story-of-autonomous-gis/</link><pubDate>Sat, 29 Mar 2025 22:30:58 -0500</pubDate><guid>https://www.huanning.me/post/2025-03-29-story-of-autonomous-gis/</guid><description>I attended the AAG Annual Meeting for the first time and gained so much from the experience. I had the chance to meet many professors, peers, and even renowned scholars whose names I had only seen in books before. One scholar I’ve long admired mentioned that our team’s article on the concept of Autonomous GIS […]</description></item><item><title>Continual geospatial agent</title><link>https://www.huanning.me/thoughts/2025-03-25-continual-geospatial-agent/</link><pubDate>Tue, 25 Mar 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/thoughts/2025-03-25-continual-geospatial-agent/</guid><description>An idea for a long-running agent with one job, such as maintaining my Street View package: adding functions, testing them, reviewing pull requests, and writing documentation.</description></item><item><title>The Zeroth Law of Geography and Geospatial Modeling</title><link>https://www.huanning.me/thoughts/2025-03-19-zeroth-law/</link><pubDate>Wed, 19 Mar 2025 00:22:25 +0000</pubDate><guid>https://www.huanning.me/thoughts/2025-03-19-zeroth-law/</guid><description>Geospatial modeling is the process of combining observations and spatial data to investigate the dynamic of phenomena over the Earth, covering physical and societal processes. Such investigations may provide explanations or predictions of phenomena (Crooks et al. 2019).</description></item><item><title>An autonomous GIS agent framework for geospatial data retrieval</title><link>https://www.huanning.me/publication/llm-find/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/publication/llm-find/</guid><description>A framework for autonomous geospatial data search and retrieval with a large language model-based agent.</description></item><item><title>GIS Copilot: towards an autonomous GIS agent for spatial analysis</title><link>https://www.huanning.me/publication/gis-copilot/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/publication/gis-copilot/</guid><description>Research toward an autonomous GIS agent for geospatial analysis.</description></item><item><title>LLM-Find: A Geospatial Data Retrieval Agent</title><link>https://www.huanning.me/project/llm-find/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/project/llm-find/</guid><description>Natural-language discovery and downloading of geospatial datasets, guided by reusable data source handbooks.</description></item><item><title>Urban Mosquito Habitat Prediction</title><link>https://www.huanning.me/project/mosquito-habitat/</link><pubDate>Wed, 01 Jan 2025 00:00:00 +0000</pubDate><guid>https://www.huanning.me/project/mosquito-habitat/</guid><description>Geospatial detection and productivity prediction for urban mosquito larval habitats to support field surveillance.</description></item><item><title>Automated floodwater depth estimation using large multimodal model for rapid flood mapping</title><link>https://www.huanning.me/publication/floodwater-depth/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://www.huanning.me/publication/floodwater-depth/</guid><description>Large multimodal model applications to flood depth estimation.</description></item><item><title>Autonomous GIS: the next-generation AI-powered GIS</title><link>https://www.huanning.me/publication/autonomous-gis-2023/</link><pubDate>Sun, 01 Jan 2023 00:00:00 +0000</pubDate><guid>https://www.huanning.me/publication/autonomous-gis-2023/</guid><description>A perspective introducing the vision of AI-powered autonomous GIS.</description></item><item><title>Converting street view images to land cover maps for metric mapping: a case study on sidewalk network extraction for the wheelchair users</title><link>https://www.huanning.me/publication/sidewalk-2022/</link><pubDate>Sat, 01 Jan 2022 00:00:00 +0000</pubDate><guid>https://www.huanning.me/publication/sidewalk-2022/</guid><description>Street-view-based land cover mapping to support sidewalk network extraction.</description></item><item><title>Exploring the vertical dimension of street view image based on deep learning: A case study on lowest floor elevation estimation</title><link>https://www.huanning.me/publication/streetview-elevation/</link><pubDate>Sat, 01 Jan 2022 00:00:00 +0000</pubDate><guid>https://www.huanning.me/publication/streetview-elevation/</guid><description>Using street-view imagery and deep learning to estimate lowest-floor elevations.</description></item><item><title>About Huan Ning</title><link>https://www.huanning.me/about/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://www.huanning.me/about/</guid><description>Computational data scientist researching autonomous GIS, geospatial AI agents, and AI for Science.</description></item><item><title>Teaching</title><link>https://www.huanning.me/teaching/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://www.huanning.me/teaching/</guid><description>Courses I have taught and supported at the University of South Carolina and New Jersey Institute of Technology, from GIS labs to AI techniques in an introductory geography course.</description></item></channel></rss>