Problem Identification

Problem Statement

⚠️

Manual monitoring in hazardous environments is unsafe, time-consuming, and inefficient

💸

Existing systems are expensive, fixed, and require continuous human control

📡

There is a need for a mobile automated system for remote surveillance and data collection

🤖

The project proposes a low-cost UGV with wireless control, sensor-based navigation, and real-time monitoring

Σ
Final Synthesis • 2026

Combined Summary: Development and Performance Validation of an Outdoor Autonomous Navigation Framework for Unmanned Ground Vehicles

Analytical Overview of 28 Research Studies (4 Members)

Key Technical Trends
  • Multi-Modal Fusion: Transition from single-sensor reliance to integrated LiDAR-Visual-Inertial-UWB pipelines.
  • Autonomous Planning: Evolution from static A* to dynamic Risk-Aware and Hierarchical Reinforcement Learning (HRL).
  • Collaborative Systems: Rise of UAV-UGV synergy for mapping and GPS-denied localization.
Performance Benchmarks

Current state-of-the-art achieves sub-10cm localization error in GPS-denied scenarios and ~97% terrain classification accuracy in complex environments like snow-covered or wooded terrains.

Primary Research Gaps

Critical bottlenecks remain in handling ultra-dynamic obstacles with low latency, distinguishing deformable vegetation from solid hazards, and maintaining communication mesh-stability in remote field deployments.

Strategic Project Value

Our project addresses the identified gaps by integrating adaptive terrain-aware path planning with a decentralized sensor fusion framework, aiming to improve UGV survival rates in unstructured off-road environments.

SYNTHESIS REMARK

"The convergence of Deep RL and Heterogeneous Multi-Robot Collaboration represents the next frontier in achieving true Level 5 autonomy for tactical and agricultural UGVs in the world's most challenging outdoor environments."

1
Sensors (MDPI) • 2020

System, Design and Experimental Validation of Autonomous Vehicle in an Unconstrained Environment

Shoaib Azam, Farzeen Munir, Ahmad Muqeem Sheri, Joonmo Kim, Moongu Jeon

Problem Addressed

Develop cost-effective full-stack autonomous vehicle using limited sensor suite for unconstrained environments.

Methodology

Four-layer architecture: Sensors → Perception → Planning → Control; Prior map-based localization + sensor fusion + rule/state-machine planning + PID/MPC control.

Sensors / Hardware

32-channel Velodyne LiDAR, 2 FLIR cameras, GNSS+IMU (Novatel), CAN bus, Raspberry Pi, server/laptop.

Dataset / Environment

KITTI dataset (training), Real-world campus/unconstrained outdoor environment.

Algorithms / Models

3D NDT localization, YOLOv3 (camera detection), PointPillars (LiDAR detection), IMM-UK-JPDAF tracking, A* path planning, State machine, PID & MPC control, Pure Pursuit.

Key Results

Accurate localization (low RMSE vs GNSS), reliable object detection, successful obstacle avoidance/stopping, validated autonomous taxi service.

Limitations

Limited sensor diversity, dependence on prior maps, proprietary vehicle control, constrained speed/testing scope.

Use for Project

Layered architecture design, NDT localization concept, LiDAR-camera fusion strategy, state-machine planning logic, PID/MPC control framework.

Unique Improvement

Enhance robustness via multi-sensor fusion (GPS+LiDAR+Visual SLAM), dynamic mapping, improved obstacle prediction, lightweight real-time models.

Strong reference for low-cost autonomous navigation systems & integrated AV stack design.
2
Drones (MDPI) • 2025

UAV Autonomous Navigation System Based on Air–Ground Collaboration in GPS-Denied Environments

Pengyu Yue, Jing Xin, Yan Huang, Jiahang Zhao, Christopher Zhang, Wei Chen, Mao Shan

Problem Addressed

Reliable UAV navigation & localization in GPS-denied, unknown environments.

Methodology

Air–ground collaborative perception; BEV image stitching + object detection + mobile UWB anchor trilateration + improved D* Lite planning.

Sensors / Hardware

DJI RoboMaster TT UAVs, downward cameras, UWB modules (Nooploop LinkTrack), laser height sensor, Wi-Fi network.

Dataset / Environment

Custom BEV image dataset; real-world outdoor experiments; GPS-denied scenario.

Algorithms / Models

SuperPoint + SuperGlue (stitching), YOLOv5 (object detection), UWB trilateration, Improved D* Lite path planning.

Key Results

Real-time global 3D mapping; <10 cm localization error; fast path planning (milliseconds); successful collision-free navigation.

Limitations

BEV coverage limitations; performance degradation under severe occlusion; global planner latency for highly dynamic obstacles.

Use for Project

Air–ground system design, BEV mapping concept, UWB localization strategy, D* Lite planning framework.

Unique Improvement

Integrate Visual SLAM + LiDAR fusion; DRL-based planner; dynamic obstacle prediction.

Strong reference for GPS-denied autonomous navigation & cooperative localization.
3
Machines (MDPI Journal) • 2018

Obstacle Avoidance System for Unmanned Ground Vehicles by Using Ultrasonic Sensors

Marco Claudio De Simone, Zandra Betzabe Rivera, Domenico Guida

Problem Addressed

Reliable obstacle detection and object recognition for UGV navigation using low-cost sensors.

Methodology

Neural-network-based object recognition using ultrasonic sensor data and MATLAB training.

Sensors / Hardware

5× SRF05 Ultrasonic Sensors, Arduino-Galileo controller, IMU sensors, DC gear motors.

Dataset / Environment

Experimental test-rig + autonomous rover tests; ~3300 sensor samples collected.

Algorithms / Models

Feed-forward Artificial Neural Network (MATLAB Neural Network Toolbox).

Key Results

Object recognition accuracy ≈77% in preliminary tests; ~99% accuracy in real rover experiments.

Limitations

Sensor interference, limited object types, small experimental environment.

Use for Project

Ultrasonic-based obstacle detection, neural network object classification approach.

Unique Improvement

Add LiDAR + camera-based perception with SLAM and advanced path planning.

Useful reference for low-cost sensor-based autonomous navigation systems.
4
Advances in Materials Science and Engineering (Wiley) • 2025

Development of a Model and Performance Analysis of the Characteristics of a Multifunctional Autonomous Unmanned Ground Vehicle

Dolya Alexandr, Kolumbetov Batyr, Kalkabek Aidos

Problem Addressed

Improving maneuverability, stability, obstacle negotiation, and autonomy of a multifunctional autonomous UGV.

Methodology

Mathematical modeling + MATLAB & RecurDyn simulations + prototype testing + modular design integration.

Sensors / Hardware

GNSS, IMU, LiDAR, Ultrasonic sensors, Infrared sensors, Cameras, NVIDIA Jetson Xavier.

Dataset / Environment

Simulation environments + Controlled real-world prototype experiments (slopes, steps, ditches).

Algorithms / Models

Kinematic & dynamic modeling, Traction force modeling, PID control, Bug 2 obstacle avoidance.

Key Results

Max climb angle ≈ 30°, obstacle height ≈ 200 mm, ditch crossing ≈ 400 mm; simulation & prototype alignment.

Limitations

Idealized assumptions (constant friction), limited test scenarios, model–prototype discrepancies.

Use for Project

Traction modeling approach, obstacle analysis methods, PID stabilization strategy, modular platform concept.

Unique Improvement

Dynamic terrain modeling + adaptive control + sensor fusion (LiDAR + Visual SLAM + GPS).

Strong reference for UGV dynamics, mobility analysis, and prototype validation.
5
HAL Preprint / Research Archive • 2025

Drone-Assisted UAV-UGV Collaboration for Autonomous Navigation in Snow-Covered Terrain

Shreyam Gupta, Pranjal Aggarwal, Priyam Gupta

Problem Addressed

Reliable navigation & localization in snow-covered, low-visibility environments.

Methodology

UAV-guided UGV navigation using road segmentation + sensor fusion + hierarchical planning + message-queue communication.

Sensors / Hardware

UAV: RGB-D Camera, GPS, IMU | UGV: IMU (minimal sensors)

Dataset / Environment

Custom UAV aerial dataset + Snow-covered simulated environment (Gazebo/ROS)

Algorithms / Models

U-Net (road segmentation), Kalman Filter (fusion), RTAB-Map (SLAM), Frontier Exploration, YOLOv5 (UGV tracking), Pure Pursuit Controller, TEB + OMPL planners.

Key Results

95.9% segmentation accuracy, 94% route success, <5% localization error, reduced path deviation (~12%), improved navigation accuracy.

Limitations

Simulation-based validation, RGB-D dependency, limited real-world testing.

Use for Project

UAV-UGV collaboration strategy, segmentation-assisted planning, hierarchical navigation, tracking + Pure Pursuit control.

Unique Improvement

LiDAR + Visual SLAM fusion, dynamic terrain adaptation, RTK-GPS / GPS-denied robustness.

Strong reference for cooperative robotics & adverse-environment navigation.
6
arXiv Preprint • 2024

A Risk-aware Planning Framework of UGVs in Off-Road Environment

Junkai Jiang, Zhenhua Hu, Zihan Xie, Changlong Hao, et al.

Problem Addressed

Safe & efficient UGV navigation in complex off-road environments considering diverse risk factors.

Methodology

Global–local planning framework integrating APF-based risk assessment + Coarse2fine A* + deterministic sampling local planning.

Sensors / Hardware

LiDAR, Cameras, Perception & Mapping modules (real-world UGV deployment).

Dataset / Environment

Simulated uncertainty maps + Real-world off-road experiments.

Algorithms / Models

Artificial Potential Field (APF), Coarse2fine A*, Deterministic Sampling, Polynomial Trajectories, QP Smoothing.

Key Results

~30% safer paths vs baseline; real-time performance; successful obstacle avoidance.

Limitations

Model assumptions, limited terrain variability, computational dependency.

Use for Project

Risk-aware planning strategy, Coarse2fine A*, APF safety modeling.

Unique Improvement

Adaptive terrain learning + sensor fusion (LiDAR + Visual SLAM + GPS).

Strong reference for safety-driven autonomous UGV navigation.
7
IJECE • 2019

Unmanned and Autonomous Ground Vehicle

S. George Fernandez et al.

Problem

Design & development of manual/autonomous UAGV for safe navigation & military applications.

Methodology

Integration of sensors + GPS navigation + Image processing + Computer vision + LabVIEW-based control.

Sensors

Camera, Ultrasonic sensors, GPS, Digital compass, IR sensors, Zigbee, DC gear motors, Arduino, LiPo battery.

Algorithms

Image processing, Obstacle detection, Shortest path (Dijkstra), GPS-based navigation.

Key Results

Successful manual & autonomous navigation; effective obstacle detection & remote control.

Use for Project

Sensor integration strategy, GPS navigation concept, obstacle detection framework.

8
ICMCSI • 2025

Design and Development of IoT Smart Driving Protocol for Unmanned Ground Vehicles

Santhosh Kumar Seelam et al.

Problem

Improve UGV navigation accuracy, responsiveness, and latency using IoT & sensor fusion.

Methodology

IoT-based modular control system with edge computing, real-time sensor fusion, adaptive decision-making.

Algorithms

Sensor fusion, path planning, obstacle avoidance, machine learning models, edge processing.

Sensors

GPS (RTK), IMU (9-axis), 360° LiDAR, ARM Cortex-M7 MCU, wireless/CAN communication.

Key Results

Improved navigation accuracy, reduced latency, enhanced responsiveness.

Use for Project

IoT modular architecture, edge computing strategy, GPS-IMU-LiDAR fusion concept.

9
APST • 2023

Recent Developments in Modular Unmanned Ground Vehicles: A Review

Abhijit Gadekar et al

Problem

Comprehensive review of modular UGV advancements, design challenges, and application trends.

Methodology

Literature survey (2016–2021); analysis of design parameters, control architectures, path planning, sensors, power systems.

Sensors

Review Paper (LiDAR, Cameras, Ultrasonic, GPS, IMU discussed).

Algorithms

Global vs Local planning, Fuzzy Logic, ANN, MPC, Genetic Algorithms, SLAM.

Key Results

Identifies trends: modularity benefits, sensor-driven autonomy, AI-based navigation improvements.

Use for Project

Modular design insights, sensor selection strategy, navigation & control taxonomy.

10
ICCCNT 2020 • 2020

Unmanned Ground Vehicle for Surveillance

Maheswaran S et al.

Problem

Short-range limitations of Bluetooth-based surveillance robots & need for safer long-range control.

Methodology

Wi-Fi controlled UGV using NodeMCU (ESP8266), Android app interface, real-time video streaming, gas sensing.

Sensors

NodeMCU (ESP8266), Wireless Camera, MQ-135 Gas Sensor, Motor Driver (L298N), DC Motors.

Algorithms

Wireless control logic, sensor monitoring (no complex AI).

Key Results

Reliable Wi-Fi control, continuous video transmission, real-time air quality monitoring.

Use for Project

Wi-Fi control architecture, NodeMCU integration, sensor interfacing strategy.

11
Alexandria Engineering Journal • 2026

Autonomous charging system via manipulator-UGV docking using zero-shot 6-DoF pose estimation

Minkyu Jung, Andrew Jaeyong Choi

Problem

Autonomous charging of UGV without human intervention and without fiducial markers.

Algorithms

YOLO-World, A*, DWA, DBSCAN, WPCA, Inverse Kinematics.

Sensors

6-DoF manipulator, TurtleBot UGV, Intel RealSense D435i, LDS-02 LiDAR, NVIDIA Jetson Xavier NX.

Key Results

90% docking success rate; average total time = 34.4 s.

Use for Project

Zero-shot detection for unseen objects; Marker-free docking; 6-DoF pose estimation pipeline.

Unique Improvement

Extend to dynamic targets; multi-robot coordination; improve alignment precision.

First integration of YOLO-World for robotic docking; marker-free & retraining-free pipeline.
12
Journal of Building Engineering • 2024

Unmanned Ground Vehicles (UGVs)-based mobile sensing for Indoor Environmental Quality (IEQ) monitoring: Current challenges and future directions

Ebrahim Alinezhad, Victor Gan, Victor W-C Chang, Jin Zhou

Problem

Fixed environmental sensors cannot capture spatial variations in indoor environmental quality (IEQ). Static sensors create undetected zones and spatial inefficiency.

Methodology

Systematic literature review of 30 selected studies on UGV-based mobile sensing. Studies categorized into UGV structure, monitoring & data capture, data analysis, and future directions.

Sensors

Mobile robots (mostly TurtleBot), sensors for temperature, humidity, CO₂, VOCs, PM2.5, illuminance, sound, localization sensors (LiDAR, cameras, encoders).

Environment

Experiments mostly in static indoor environments such as universities, labs, factories, and schools.

Algorithms

SLAM, Kalman Filter, interpolation (Kriging, cubic, nearest neighbor), machine learning models such as SVR, KNN, Linear Regression, Random Forest, Decision Trees.

Key Results

UGV sensing provides high-resolution spatial-temporal environmental data, detects pollutant hotspots, improves IEQ mapping, and can reduce building energy consumption by ~3%.

Limitations

Sensor response time (SRT) errors, localization challenges, limited real-world testing, limited validation methods, difficulty handling dynamic human occupancy.

Use for Project

Use mobile robots with environmental sensors to collect spatial environmental data; combine with machine learning models to predict indoor environmental conditions.

Unique Improvement

Integrate 3D indoor mapping, improved localization, optimized robot velocity for sensing, and integration with Building Automation Systems (BAS).

Mobile sensing can significantly improve indoor environmental monitoring compared to fixed sensors.
13
IFAC PapersOnLine • 2025

Developing a Communication Node Deployment System for UGVs

C. Fisher, M.J. McDermott

Problem

Robots operating in underground mines or disaster environments often lose communication due to lack of network infrastructure. This prevents real-time data transfer during search and rescue missions.

Methodology

Developed a UGV-mounted system that deploys wireless communication nodes (breadcrumbs) to create a temporary mesh network while the robot moves through the environment.

Sensors

UGV platform (Clearpath Husky A200), ESP32 microcontrollers, servo motors, accelerometer sensors, N20 motor with lead screw, Wi-Fi antennas.

Environment

Tested in uneven terrain environments simulating subterranean mining or SAR scenarios.

Algorithms

ESP-NOW communication topology, accelerometer-based orientation control, servo control loop.

Key Results

System successfully deployed nodes and created communication extension capability; servo-actuated node self-righting time ≈1.9 s and gravity-gimbal node ≈2.86 s.

Limitations

Prototype-level system, limited environmental testing, mesh network performance not fully evaluated, communication modules may need improvement for harsh environments.

Use for Project

Deployable mesh communication nodes for UGV exploration; modular node deployment architecture; self-righting communication node concept.

Unique Improvement

Add automatic node placement optimization using signal strength or AI-based positioning, integrate long-range communication (LoRa / UWB).

Useful for SAR robots, mining robots, and exploration robots where network infrastructure is unavailable.
14
IEEE IWCMC • 2025

Adaptive P2P UAV-UGV Prototype for Ad Hoc Network-based Precision Agriculture

Demirel Memet, Ionuț Pirnog, et al.

Problem

Precise communication for autonomous systems in agriculture often relies on 4G/5G, which is limited in vast/inaccessible terrains, hindering data relay to servers.

Methodology

Development of a P2P ad hoc network using UAVs as mobile access points. Integrates Smell Agent Optimization (SAO-P2PS) for dynamic peer selection under fluctuating link conditions.

Sensors

UGV (environmental sensors for soil/humidity), UAV (mobile AP), Wi-Fi/Bluetooth communication modules.

Environment

Vast agricultural terrains with limited or zero 4G/5G coverage.

Algorithms

Smell Agent Optimization-based Peer-to-Peer Selection (SAO-P2PS) algorithm.

Key Results

SAO-P2PS achieves 25-35% higher throughput and 40-60% better energy efficiency compared to Random or Nearest-Neighbor selection.

Limitations

Communicational network instability, computational constraints, and environmental adaptability in extreme weather.

Use for Project

Cost-efficient ad hoc communication deployment; minimized reliance on infrastructure; real-time data collection framework.

Unique Improvement

Circumvents the need for traditional 4G/5G via adaptive UAV relay; optimized P2P selection for energy/throughput.

15
Computers and Electronics in Agriculture • 2025

Tactile Sensing & Visually-Impaired Navigation in Densely Planted Row Crops

Adam M. Gronewold, Philip Mulford, Eliana Ray, Laura E. Ray

Problem

Traditional navigation (GNSS/Vision) fails in dense row crops due to signal attenuation, multi-path errors, and leaf-induced visual clutter.

Methodology

Fusion of tactile mechanical probes with LiDAR. Filters rigid obstacles (stalks) from flexible features (leaves) to navigate without visual input.

Sensors

Mechanical probe with rotary encoder, 2D/3D LiDAR (Hokuyo UTM-30LX/Velodyne VLP-16), IMU, Wheel encoders.

Environment

Densely planted sorghum and corn fields with narrow row spacing and cluttered canopies.

Algorithms

Obstacle mapping, heuristic row-line estimation, linear least squares for within-row distance, Model Predictive Control (MPC).

Key Results

97% obstacle detection accuracy; 4cm global positioning accuracy; successfully navigated >100m in simulation and 30m in real fields.

Limitations

High computational complexity of MPC solvers; reliance on rigid vs flexible feature differentiation; sensitivity to extreme row curvature.

Use for Project

Vision-independent navigation framework; tactile-LiDAR fusion strategy; robust under-canopy operation.

Unique Improvement

Explicitly differentiates rigid (stalks) vs flexible (leaves) obstacles; vision-independent stability in dense forest/crop environments.

16
HardwareX (Elsevier) • 2021

Cable Deployment System for Unmanned Ground Vehicle (UGV) Mobile Microgrids

John E. Naglak, Caleb Kase, Max McGinty, Casey D. Majhor, Carl Greene, Jeremy Bos, Wayne Weaver

Problem

Autonomous microgrids require physical cable connections between distributed energy nodes, but manual cable deployment is inefficient and dangerous in complex environments.

Methodology

Designed a custom Adjustable Cable Management Mechanism (ACMM) that automatically deploys and retracts cable while the UGV moves, preventing slack or entanglement.

Sensors

UGV platform, NEMA 24 stepper motor, Arduino Uno, rotary encoder, motor driver (DMT542T), cable reel system.

Environment

Outdoor terrain demonstration with a UGV connecting power source and load nodes.

Algorithms

Feedforward + PID feedback control based on UGV velocity and cable angle estimation.

Key Results

Precise cable control ±1 inch over 70 ft cable, adjustable speed up to 3.3 ft/s, lightweight (~5 lbs system).

Limitations

3D printed components may lack durability for long-term use; limited real-world testing; requires proper calibration.

Use for Project

Cable deployment mechanism, velocity-based cable control, Arduino + stepper motor implementation.

Unique Improvement

Use AI-based path prediction to optimize cable route, integrate autonomous cable obstacle detection.

Designed as a low-cost open-source system (~$434) for robotic applications.
17
Sensors (MDPI) • 2025

Robust and Precise Navigation and Obstacle Avoidance for Unmanned Ground Vehicles

González-Hernández et al.

Problem

UGVs face traction loss and irregular terrain; need robust trajectory tracking without complex derivatives.

Methodology

Simplified 2nd-order SMC; Cascaded architecture (trajectory gen, pose estimation, LiDAR evasion, speed regulation).

Sensors

RPLiDAR A2, GNSS, redundant IMUs, Magnetometer; 4WD differential-drive UGV.

Algorithms

Second-order Sliding Mode Control (SMC), GNSS-IMU pose fusion, LiDAR point cloud processing.

Key Results

Tracking error ~0.08m; std. dev. of position/velocity low; robust to disturbances.

Use for Project

SMC for robust outdoor control; LiDAR-GNSS pose fusion; real-time evasion logic.

18
Scientific Reports (Nature) • 2026

Autonomous navigation in unstructured outdoor environments using hierarchical deep reinforcement learning

Ahmed Tibermacine et al.

Problem

UGVs struggle with unstructured outdoor navigation due to unpredictable obstacles and terrain variations.

Methodology

Hierarchical deep RL: Semantic perception + High-level RL policy + Low-level geometric controller.

Sensors

RGB-D cameras, LiDAR; wheeled/tracked UGV; NVIDIA GPU.

Algorithms

HRL (PPO/DDPG), Semantic segmentation CNN (U-Net/DeepLab), Reward shaping.

Key Results

86.7% trail-following success; low collision frequency; precise path tracking; superior vs. baselines.

Use for Project

Hierarchical RL architecture; semantic perception pipeline; reward shaping strategies.

19
Sensors (MDPI) • 2023

Reinforcement and Curriculum Learning for Off-Road Autonomous Navigation of an UGV

Manuel Sánchez, et al.

Problem

Off-road navigation requires handling unstructured terrain traversability without extensive generic training data.

Methodology

Deep RL with Actor-Critic; Curriculum Learning across 3 Gazebo scenarios; 3D LiDAR → virtual 2D traversability scanner.

Sensors

3D LiDAR, Andabata UGV platform, onboard computing.

Algorithms

Actor-Critic RL, Random Forest traversability classifier.

Key Results

Successful sim-to-real transfer; collision avoidance and goal reaching in natural terrain; accelerated convergence.

Use for Project

Traversability mapping from 3D LiDAR; curriculum learning; Actor-Critic RL.

20
CoDIT • 2016

An integration framework for UGV outdoor navigation system based on multi-sensor fusion

Lazhar Khriji, et al.

Problem

Seamless integration of separate modules (perception, planning, control) for real-time outdoor navigation.

Methodology

Three-module architecture: Line/obstacle detection, Path planning/tracking, Real-time motor control.

Sensors

LiDAR scanner, monocular camera, motor control system hardware.

Algorithms

LiDAR-camera fusion, path planning within boundaries, motor control.

Key Results

Successful real-time navigation within boundaries; demonstrated system integration performance.

Use for Project

Modular three-layer architecture; LiDAR-camera fusion pipeline; integration framework.

21
Virginia Tech Dissertation • 2025

Terrain-Aware Tactical Motion Planning and Control of Off-Road Unmanned Ground Vehicles

Jyotirmoy Mukherjee

Problem

UGVs in GNSS-denied terrains need terrain-aware stealth navigation stack without GPS.

Methodology

Hierarchical stack: vision-inertial mapping (VIO) → stealth planning → trajectory optimization → control.

Sensors

Vision sensors, IMUs, Intel RealSense, Clearpath Husky A200.

Algorithms

Vision-Inertial Odometry (VIO), terrain segmentation, trajectory optimization, robust control.

Key Results

Successful GNSS-denied navigation; 28Hz terrain segmentation; validated indoor/outdoor tracking.

Use for Project

GNSS-denied VIO pipeline; hierarchical autonomy stack; terrain-aware planning.

22
IEEE TVT • 2025

Real-Time Navigation of Unmanned Ground Vehicles in Complex Outdoor Environments

Han, Chen et al.

Problem

Precise real-time maneuvering across challenging outdoor terrains with dynamic obstacles and uncertainty.

Methodology

Multi-modal sensor fusion + enhanced perception + memory-guided decision-making strategies.

Sensors

LiDAR, Cameras, IMU, GNSS; high-performance onboard computing.

Algorithms

Deep learning fusion, memory-augmented navigation policies, adaptive trajectory optimization.

Key Results

Superior real-time performance; robust decision-making; validated field performance.

Use for Project

Multi-modal sensor fusion pipeline; memory-guided navigation strategies; terrain classification.

23
PERMIS • 2010

Evaluating the Performance of Unmanned Ground Vehicle Water Detection

Shane Brennan, Alberto J. Rodriguez, et al.

Problem

Reliable water hazard detection for cross-country navigation; lack of standardized metrics.

Methodology

Four image-space water detection algorithms fused into terrain classification; stereo ranging in 3D maps.

Sensors

Stereo camera pairs (3 baselines), GDRS ladar, IMU, GPS; experimental XUVs.

Algorithms

Water detection fusion, stereo range fusion, temporal filtering in terrain maps, Kalman-filtered GPS/IMU.

Key Results

Established detection range/accuracy metrics; ground truth within 0.5% positioning error.

Use for Project

Water detection fusion methodology; image vs. map space metrics; ground-truth validation protocol.

24
Transportation Research Procedia • 2021

Optimal path planning of autonomous navigation in unstructured outdoor environment

S. Julius Fusic, G. Kanagaraj, K. Hariharan, S. Karthikeyan

Problem

Collision-free global path planning in unstructured environments using satellite imagery without ground sensors.

Methodology

Satellite image processing for path extraction + heuristic cost map generation + optimal planning algorithm.

Sensors

Satellite imagery (offline/remote source).

Algorithms

Satellite image segmentation, heuristic cost functions, A*-variant optimal path search.

Key Results

Generates paths 3-5x faster than traditional A*; 95%+ path optimality on unstructured terrains.

Use for Project

Satellite preprocessing as global planner; heuristic cost map generation module.

25
Smart Agricultural Technology • 2025

Development of a Stereo Vision-Based UGV Guidance System for Bareroot Forest Nurseries

Sharif Shabani, Ashish R. Mulaka, Thomas A. Stokes, Tanzeel U. Rehman, Yin Bao

Problem

Manual forest nursery inventory and navigation is labor-intensive and inefficient. Existing navigation methods (GNSS, LiDAR) are expensive or unreliable in agricultural environments.

Methodology

Developed an automatic guidance system for a UGV using stereo vision and depth image processing to detect nursery bed edges and guide the vehicle.

Sensors

Skid-steering UGV, ZED 2i stereo camera, RTK-GNSS receiver, Jetson AGX Orin edge computer, LED lighting, OAK-D stereo cameras.

Environment

Tested in Webots simulation and real-world forest nursery fields with curved seedling beds.

Algorithms

Depth image processing, Canny edge detection, Hough line transform, proportional control steering algorithm.

Key Results

Successfully completed 1422 m navigation without collisions, average RMS deviation ≈ 0.05 m from bed centerline.

Limitations

Performance affected by lighting conditions, bed geometry, and soil conditions; stereo camera depth noise and detection failures at higher speed.

Use for Project

Stereo vision navigation, ROS-based robotic control architecture, depth-image processing for path detection.

Unique Improvement

Improve robustness using deep learning segmentation, adaptive controllers, and multi-sensor fusion (LiDAR + vision).

Demonstrates low-cost autonomous navigation for agricultural robots.
26
Studia Informatica • 2025

Autonomous Navigation for Unmanned Ground Vehicles in Unstructured Terrain

M. Nowakowski

Problem

UGVs struggle with perception and traversability in unstructured environments like forests and mountains.

Methodology

Presents TAERO platform; reviews perception challenges and AI solutions for unstructured operations.

Sensors

LiDAR, cameras, radar, UWB radar for vegetation penetration; TAERO UGV platform.

Algorithms

Terrain classification (SPOC CNN), deep learning sensor interpretation, real-time path planning.

Key Results

TAERO demonstrated for reconnaissance; UWB radar distinguishes grass/bushes from solid obstacles.

Use for Project

TAERO platform requirements as benchmark; UWB radar vegetation penetration concept.

Identifies exact gaps (decision-making under uncertainty) that our framework addresses.
27
Results in Engineering (Elsevier) • 2025

OS-RFODG: Open-source ROS2 framework for outdoor UAV dataset generation

Imen Jarraya, Mohamed Abdelkader, Khaled Gabr, Muhammad Bilal Kadri, Fatimah Alahmed, Wadii Boulila, Anis Koubaa

Problem

Existing UAV datasets lack sensor synchronization, terrain awareness, and realistic environmental conditions, making localization research difficult.

Methodology

Proposed OS-RFODG, a ROS2-based framework that integrates QGIS, Blender, Gazebo, PX4, and QGroundControl to generate synchronized multi-sensor UAV datasets with realistic terrain models.

Sensors

LiDAR, GPS, IMU, RGB Camera, Barometer.

Environment

Simulation environment based on Gazebo + PX4 SITL, terrain from GeoTIFF elevation maps (Makkah region, Saudi Arabia).

Algorithms

Sensor fusion, Visual localization (SIFT feature matching), trajectory analysis, dataset benchmarking.

Key Results

RMSE for trajectory elevation: 13.59–19.5 m; image localization precision up to 0.8964; visual localization error < 1.34 m.

Limitations

Simulation-based dataset; elevation errors increase in complex terrain; environmental disturbances like wind/rain not fully modeled.

Use for Project

Framework pipeline for multi-sensor dataset generation, ROS2 synchronization techniques, terrain modeling pipeline using DEM/DSM + Blender.

Unique Improvement

Add real UAV flights + real sensor noise models, integrate AI-based localization, include dynamic environmental disturbances.

Very useful for GNSS-denied UAV localization research and dataset generation.
28
Robotics and Autonomous Systems (Elsevier) • 2024

3D Traversability Analysis and Path Planning Based on Mechanical Effort for UGVs in Forest Environments

Afonso E. Carvalho, João Filipe Ferreira, David Portugal

Problem

Autonomous navigation in rough forest terrains is difficult because robots must avoid obstacles, steep slopes, and unsafe paths while minimizing energy consumption.

Methodology

Proposed Mechanical Effort Based Traversability (MEBT) framework that analyzes 3D point cloud terrain maps, computes terrain gradient and mechanical effort, and generates an optimal path on a traversability costmap.

Sensors / Hardware

UGV robot platform (Clearpath Husky), Velodyne LiDAR, IMU, wheel encoders.

Environment

Gazebo simulation scenarios and real forest dataset (Montmorency Forest, Canada).

Algorithms

Gradient computation, traversability analysis, mechanical effort estimation, A* global planning with ROS move_base.

Key Results

Generated safer paths with lower mechanical effort and lower pitch risk; significantly improved map processing speed compared to previous mechanical-effort approaches.

Limitations

System relies on pre-mapped point cloud environment; limited handling of dynamic obstacles; planning mainly focuses on global planning rather than full autonomy.

Use for Project

Traversability map generation, gradient-based terrain evaluation, mechanical-effort path planning concepts.

Unique Improvement

Add real-time perception + deep learning terrain classification + adaptive path planning using reinforcement learning or MPC.

Useful for UGV navigation in forests, agriculture, or rough terrain robotics.
1. Existing Products in Market
ProductCompanyApplication Navigation / SensingPayload / Mobility Approx. PriceRemarks
Husky A200 UGVClearpath RoboticsOutdoor R&D, inspectionROS-ready, LiDAR/GPS/IMU75 kg, all-terrain₹16.8 lakhBase price listed
Husky A300 UGVClearpath RoboticsAutonomous roboticsROS 2 compatible100 kg, IP54Not disclosedCommercial quote
RB-VOGUIRobotnikInspection, logisticsAutonomous, sensor integrationIndustrial baseNot disclosedVendor pricing
RB-FIQUSRobotnikHeavy industrial useModular autonomous systemHeavy payloadNot disclosedEnterprise model
HK1000-V2SuperDroid RobotsAutonomous robot devGPS + LiDAR supportHeavy-duty terrain₹8–15+ lakhConfig-based
Proposed UGVAcademic PrototypeNavigation & surveillanceGPS L76x + IMU + LiDAR + ROSCompact₹85k – ₹1.15 lakhStudent model
2. Key Market Observations
Most UGVs are designed for:
  • Research
  • Industrial inspection
  • Outdoor autonomous operations
Common Features:
  • Autonomy-ready software
  • Sensor integration
  • Rugged mechanical design
Major Limitations:
  • High cost
  • Large size
  • Overkill for academic use
3. Market Gap Identified
  • Low-cost solutions
  • Compact design
  • Easy integration
  • Affordable modularity
  • Student-friendly deployment
👉 Gap: A low-cost, compact, ROS-based outdoor UGV suitable for academic and research use.
4. Proposed System Positioning
  • Low-cost solution
  • Outdoor + semi-structured operation
  • Real-time monitoring
  • Sensor fusion (GPS + IMU + LiDAR)
  • ROS-enabled scalability
5. Competitive Advantage
FeatureExisting UGVsProposed UGV
CostHighLow
PlatformIndustrialAcademic
Outdoor CapabilityYesYes
ModularityMedium–HighHigh
ReplicationDifficultEasy
Student UseLimitedStrong
Sensor FusionYesYes
1. Existing Patents
Patent No.TitleYearCore IdeaTechnologies Used
US12179737B2Unmanned Ground Vehicle and Method for Operating UGV2024Waypoint-based navigation with obstacle replanningWaypoint Navigation, Obstacle Detection
US12269720B2Ground-Based Transport Vehicle2025Autonomous transport with minimal human inputNavigation Logic, Collision Avoidance
US10240930B2Sensor Fusion2019Combines multiple sensors for accurate localizationSensor Fusion, Data Integration
US12019173B2Lane-Level Navigation using LiDAR & Signals2024High-precision navigation using LiDAR + signalsLiDAR Navigation, Localization
US11573090B2LiDAR and Remote Localization2023LiDAR-based mapping and navigationLiDAR Mapping, Obstacle Detection
2. Key Patent Observations
Strong focus on:
  • Autonomous navigation
  • LiDAR-based localization
  • Sensor fusion
  • Real-time obstacle handling
Most systems designed for:
  • Industrial-grade deployment
  • Commercial autonomous platforms
Common Characteristics:
  • High computational requirements
  • Expensive sensor stacks
  • Complex integration
3. Patent Gap Identified
  • Low-cost implementation
  • Student-scale deployment
  • Simple ROS-based systems
  • Affordable outdoor UGV architecture
👉 Gap: Need for a cost-effective, simplified autonomous UGV system suitable for academic and small-scale applications.
4. Proposed Improvements
  • Efficient GPS + IMU + LiDAR fusion
  • Lightweight and scalable architecture
  • Real-world outdoor usability
  • Low-cost implementation