Share

CHAN LOG

SECTION

Computer Graphics

Machine Learning

Computer Vision

Prompt Engineering

NEWEST

Newest

Search

A Survey on Robotics with Foundation Models: toward Embodied AI

A Survey on Robotics with Foundation Models: toward Embodied AI

While the exploration for embodied AI has spanned multiple decades, it remains a persistent challenge to endow agents with human-level intelligence, including perception, learning, reasoning,...

https://arxiv.org/abs/2402.02385

A survey by Midea Group researchers systematically reviews the application of foundation models in robotics, focusing on autonomous manipulation, by categorizing approaches for high-level planning and low-level control, and identifying key challenges and future research directions in embodied AI.

Understanding Foundation Models in Robotics

The field of robotics stands at a transformative juncture where powerful artificial intelligence models, originally designed for processing internet-scale text and images, are being adapted to control physical robots in the real world. This comprehensive survey by Xu et al. examines how Foundation Models (FMs) - including Large Language Models (LLMs), Large Vision Models (LVMs), and Vision-Language Models (VLMs) - are revolutionizing robotics by enabling more intelligent, adaptable, and generalizable robotic systems.

Foundation Models represent a paradigm shift from traditional robotics approaches. Unlike conventional robots that operate based on pre-programmed actions in controlled environments, FMs enable robots to understand natural language commands, reason about complex scenarios, and adapt to new situations. These models, pre-trained on vast datasets from the internet, possess remarkable capabilities in understanding, reasoning, interaction, and generation that can be leveraged for embodied AI applications.

The Dual Framework of Robot Intelligence

The authors organize their analysis around two fundamental aspects of robotic intelligence: high-level planning and low-level control. This framework mirrors how humans approach complex tasks - first deciding what to do (planning) and then executing those decisions with precise movements (control).

High-Level Planning with Foundation Models

High-level planning involves the strategic decision-making process where robots interpret commands and develop step-by-step plans for complex tasks. The survey identifies two primary categorizations for how FMs contribute to planning:

Forms of Planning refers to the output format that FMs generate for planning purposes. The authors identify three main approaches:

•

Structured Language Planning: Uses formal languages like Planning Domain Definition Language (PDDL) to generate precise, unambiguous plans

•

Policy Code Planning: Generates executable code that serves as the robot's action policy

•

Natural Language Planning: Produces flexible, human-readable plans in natural language

A Survey on Robotics with Foundation Models: toward Embodied AI

Embodied AI

ENG

Dijkstra Algorithm

개요

다익스트라 알고리즘

다익스트라 알고리즘의 기본 동작 방식은 다음과 같다:

시작 노드 설정: 출발점이 되는 시작 노드를 선택하고, 그 노드에서 다른 노드로 가는 경로를 계산한다. 처음에는 시작 노드에서 자신의 거리를 0으로 설정하고, 다른 모든 노드까지의 거리는 무한대로 초기화한다.

방문하지 않은 노드 중에서 최단 거리를 가진 노드 선택: 시작 노드에서 출발하여 인접한 노드 중 가장 가까운 노드를 선택하고 그 노드를 탐색한다. 아직 방문하지 않은 노드들 중에서 현재까지의 최단 거리가 가장 짧은 노드를 선택한다.

최단 거리 갱신: 선택한 노드의 인접한 노드를 확인하고, 해당 노드를 거쳐 가는 경로가 더 짧으면 그 노드까지의 거리를 갱신한다.

이 과정을 반복: 모든 노드를 탐색할 때까지 위의 과정을 반복하며, 한 번 방문한 노드는 다시 방문하지 않는다. 각 노드를 탐색할 때마다 우선순위 큐를 사용하여 가장 작은 값을 가진 노드를 효율적으로 선택한다.

이 과정이 끝나면, 시작점에서 각 노드까지의 최단 경로가 결정된다.

Dijkstra Algorithm

From the Bottom

Handling CPU-bound Tasks in FastAPI

Using ThreadPoolExecutor + uvicorn --workers for Effective Parallel Processing

Background

FastAPI is known for its powerful asynchronous capabilities, especially with I/O-bound tasks. However, when it comes to CPU-bound workloads (like OCR, image processing, or heavy computation), async def alone doesn't bring much performance gain.

This post walks you through how I improved the responsiveness and concurrency of my FastAPI app using:

ThreadPoolExecutor for CPU-bound operations

Uvicorn’s -workers option for true multi-processing

The Problem

While building an API for generating scripts using LLMs, I ran into performance bottlenecks due to:

Intensive OCR and image preprocessing (e.g., with PyMuPDF, OpenCV)

Running dozens of LLM summarization and generation calls

Handling large text files and saving structured data

Dev Log [05.29.25] : Handling CPU-bound Tasks in FastAPI

OVERVIEW

All of my posts are below. Hope you can find what you need :D
Search on the database view below, and you will get what you are looking for!

Gallery

Search

XR

2

Meta XR SDK: 1) Ray 기반 UI Button 구현

Meta Interaction SDK

XRI - 1) Scene Setup

XR Interaction Toolkit

Unity

5

GPP - Part 1. GoF’s Design Patterns - 3) Observer Pattern

Game Programming Pattern

GPP - Part 1. GoF’s Design Patterns - 2) Flyweight Pattern

Game Programming Pattern

GPP - Part 1. GoF’s Design Patterns - 1) Commander Pattern (C#, Unity)

Game Programming Pattern

GPP - Part 1. GoF’s Design Patterns - 1) Commander Pattern

Game Programming Pattern

GPP - Part 0. Preface

Game Programming Pattern

Prompt Enginnering

6

Ollama 특정 GPU만 사용하기 설정 방법

RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval

Long Context Search

What is Vector Similarity Search? : 2. How Does Vector Similarity Work

Vector Similiarty Search

Vector Embedding

What is Vector Similarity Search? : 1. Introduction

Vector Similiarty Search

LLM Agent Overview

Langchain: Pydantic

ML

6

기초 정보 이론 for ML

Information Theory

What is the FID?

What is an “Instruct Model” in LLM?

PDF를 OCR로 전처리하기

Ollama Template 파헤치기 - Go Template 문법을 중심으로

Colab 으로 Llama 3 파인튜닝하기 (feat. Unsloth, Ollama)

Life Log

4

이나모리 가즈오, 부러지지 않는 마음

Self-Motivation

Ultra-Learning On My Mind

서울대 IPAI 대학원 구술 면접

샘 알트먼의 13가지 성공의 법칙

CSE

7

데이터통신의 전송기술

데이터통신의 회선 구성

전화망의 이해

인터넷과 인터넷 주소체계

OSI 참조 모델

프로토콜의 주요 요소 및 기능

데이터 통신의 이해

Algorithm

17

Dijkstra Algorithm

From the Bottom

Priority Queue (feat. Heap)

From the Bottom

Binary Search (feat. Binary Search Tree)

Search Algorithm

From the Bottom

Basics of Dynamic Programming

AI 대학원 기출문제

From the Bottom

Topological Sorting

From the Bottom

Graph - Breadth First Search

From the Bottom

Graph - Depth First Search

From the Bottom

Introduction to Graph Algorithm

From the Bottom

Comparison of Sorting Algorithm (feat. External Sort)

Sorting Algorithm

From the Bottom

Sorting Algorithm - Quick Sort (쾌속 정렬)

Sorting Algorithm

From the Bottom

Sorting Algorithm - Merge Sort (합병 정렬)

Sorting Algorithm

From the Bottom

Sorting Algorithm - Shell Sort (쉘 정렬)

Sorting Algorithm

From the Bottom

Sorting Algorithm - Insertion Sort (삽입 정렬)

Sorting Algorithm

From the Bottom

Sorting Algorithm - Bubble Sort (버블 정렬)

Sorting Algorithm

From the Bottom

Sorting Algorithm - Selection Sort (선택 정렬)

Sorting Algorithm

From the Bottom

From the Bottom

Introduction to From the Bottom - Algorithm Series

Personal Statement

From the Bottom

1개 그룹 더 표시

DEV LOG

Here are my logs as a developer.

Section

Search

Basic

7

Dev Log [01.14.2025] : [Langgraph Chat Memories]

Dev Log [01.14.2025] : [Langgraph Chat Memories]

Dev Log 05.26.24 : Unity UI Sys. to Quest

Meta Interaction SDK

Dev Log 05.26.24 : Unity UI Sys. to Quest

Meta Interaction SDK

Dev Log 02.18.24 : Python Object Reference

Dev Log 02.18.24 : Python Object Reference

Dev Log 02.15.24 : Python Print Text Color

Dev Log 02.15.24 : Python Print Text Color

Dev Log 02.14.24 : Python Mutable Obejct & Shallow Copy

Dev Log 02.14.24 : Python Mutable Obejct & Shallow Copy

Dev Log: 01.17.24: LangChain ChromaDB

Dev Log: 01.17.24: LangChain ChromaDB

Dev Log 01-03-24: Python File RW

Dev Log 01-03-24: Python File RW

Trouble Shooting

8

Dev Log [03.05.25] : PyTorch Integer Type Conversion

Dev Log [03.05.25] : PyTorch Integer Type Conversion

Dev Log 03.04.24 : Object of type 'FieldInfo' is not JSON serializable

Dev Log 03.04.24 : Object of type 'FieldInfo' is not JSON serializable

Dev Log 03.03.24 : Pydantic Error - ConfigError: unable to infer type for attribute "introduction"

Dev Log 03.03.24 : Pydantic Error - ConfigError: unable to infer type for attribute "introduction"

Dev Log 02.28.24 : Quest3 VR Application Trouble Shooting

Meta Interaction SDK

Dev Log 02.28.24 : Quest3 VR Application Trouble Shooting

Meta Interaction SDK

Dev Log 02.16.24 : Pydantic Error : Field Required

Dev Log 02.16.24 : Pydantic Error : Field Required

Dev Log 02.15.24 : VS Code Python Interpreter Setting

Dev Log 02.15.24 : VS Code Python Interpreter Setting

Dev Log 01.23.24 : Git Error - src refspec master does not match any

Dev Log 01.23.24 : Git Error - src refspec master does not match any

Dev Log 12.26.23 : Git Rebase Drop

Dev Log 12.26.23 : Git Rebase Drop

Project

1

Dev Log 01.27.24 : My First Watch Dog

Dev Log 01.27.24 : My First Watch Dog

Log

7

Dev Log [05.29.25] : Handling CPU-bound Tasks in FastAPI

ThreadPoolExecutor

Dev Log [05.29.25] : Handling CPU-bound Tasks in FastAPI

ThreadPoolExecutor

Dev Log [02.11.2025] : Pydantic 모델에서 Union(|) 연산자로 상속 문제 해결하기

Dev Log [02.11.2025] : Pydantic 모델에서 Union(|) 연산자로 상속 문제 해결하기

Dev Log [02.07.25] : Unity Event 오랜만에..

Dev Log [02.07.25] : Unity Event 오랜만에..

Dev Log [01.14.2025] : [Langgraph Chat Memories]

Dev Log [01.14.2025] : [Langgraph Chat Memories]

Dev Log 05.17.24 : Python Method Overloading

Dev Log 05.17.24 : Python Method Overloading

Dev Log 02.12.24 : Meta XR Simulator

XR Interaction Toolkit

Dev Log 02.12.24 : Meta XR Simulator

XR Interaction Toolkit

Dev Log 02.01.24 : Tiny Experience on SW Requirement..

Dev Log 02.01.24 : Tiny Experience on SW Requirement..

Programming

4

Dev Log 05.28.24 : 유니티 한글 입력 로직 구현

Dev Log 05.28.24 : 유니티 한글 입력 로직 구현

Dev Log 05.26.24 : Unity UI Sys. to Quest

Meta Interaction SDK

Dev Log 05.26.24 : Unity UI Sys. to Quest

Meta Interaction SDK

Dev Log 03.24.24 : Draw Outline of Selected Objects

Dev Log 03.24.24 : Draw Outline of Selected Objects

Dev Log 02.10.24 : Initializer of Constructor..?

Dev Log 02.10.24 : Initializer of Constructor..?

Code

6

Dev Log [03.05.25] : PyTorch Integer Type Conversion

Dev Log [03.05.25] : PyTorch Integer Type Conversion

Dev Log [02.11.2025] : Pydantic 모델에서 Union(|) 연산자로 상속 문제 해결하기

Dev Log [02.11.2025] : Pydantic 모델에서 Union(|) 연산자로 상속 문제 해결하기

Dev Log [02.04.25] : Open Source Contribution; Bringing BERTScore with score_with_vectors

Dev Log [02.04.25] : Open Source Contribution; Bringing BERTScore with score_with_vectors

Dev Log 05.28.24 : 유니티 한글 입력 로직 구현

Dev Log 05.28.24 : 유니티 한글 입력 로직 구현

Dev Log 05.17.24 : Python Method Overloading

Dev Log 05.17.24 : Python Method Overloading

Dev Log 03.09.24 : UI Rotation Following the Camera

Dev Log 03.09.24 : UI Rotation Following the Camera

Thank you for visiting!
Written by Changyu Lee