MiniUGV$_2$: A Compact UAV-Deployable Tracked Ground Vehicle with Manipulation Capabilities

Durgakant Pushp; Swapnil Kalhapure; Shaekh Mohammad Shithil; Lantao Liu

cs.ROMar 1, 2026

MiniUGV $_2$ : A Compact UAV-Deployable Tracked Ground Vehicle with Manipulation Capabilities

Authors: Durgakant Pushp, Swapnil Kalhapure, Shaekh Mohammad Shithil, Lantao Liu

Organizations: Luddy School of Informatics, Computing, and Engineering at Indiana University, Bloomington, IN 47408, USA. · S. Kalhapure is an Independent Researcher at the time of work.

Abstract

Exploring and inspecting Hidden Spaces, defined as environments whose entrances are accessible only to aerial robots but remain unexplored due to geometric constraints, limited flight time, and communication loss, remains a major challenge. We present miniUGV $_2$ , a compact UAV-deployable tracked ground vehicle that extends UAV capabilities into confined environments. The system introduces dual articulated arms, integrated LiDAR and depth sensing, and modular electronics for enhanced autonomy. A novel tether module with an electro-permanent magnetic head enables safe deployment, retrieval, and optional detachment, thereby overcoming prior entanglement issues. Experiments demonstrate robust terrain navigation, self-righting, and manipulation of objects up to 3.5 kg, validating miniUGV $_2$ as a versatile platform for hybrid aerial-ground robotics.

Figures & tables

Figure 1: Illustration of the UAV-miniUGV system: Left and right images depict the designed miniUG $V_{2}$ , an improved version of the miniUGV Pushp et al. (2022) . Center image shows the primary and secondary robots together with a tether module.

Figure 2: Proposed system architecture of the UAV-miniUGV platform for collaborative exploration. The UAV is equipped all the entire sensor suite to create a 3D mapping of the environment and find potential deployable zones for the miniUGV. The UAV then autonomously deploy and retrieve the robot using magnetic tether head. The miniUGV has articulated arms for manipulation and navigation in rough terrains. Both robots share their trajectory information via WiFi network and the high level mission command is sent from the Base station.

Figure 3: Overview of the tether head position tracking.

Panel of Figure 4 — Figure 4: (a) Assembled miniUG $V_{2}$ , showcasing its features, highlighting its compact size with integrated sensors and gripper mechanism. (b) Shows the designed circuit board with all the electronic components required to operate the miniUG $V_{2}$ .

Panel of Figure 5 — Figure 5: (a) Shows the the tether module design on the left side and the actual prototype (3D printed using the red color PLA material) with the tether control onboard computer on the right side.(b) Shows the magnetic tether head. It is a fully autonomous ROS-enabled module incorporating an ESP32 microcontroller, a LiPo battery for power supply, and a voltage regulator that distributes power to both 5V and 3.3V rails.

No.	Robot	Dimension L/W/H(mm)	Weight (kg)	Operation Time(hrs)	Sensors, actuators & Computer	Gripper	Active Flippers
1	EyeDrive [ 9 ]	280/250/100	2.3	3.0	1. 5 Cameras, 2. IMU, 3. Microphone Remote- Controlled	NA	No
2	miniUG $V_{1}$ Pushp et al. (2022)	130/120/55	$<$ 0.25	0.3	1. RGB Camera, 2. IMU, Autonomous- RPi Zero	Magnetic	No
3	miniUG $\mathbf{V_{2}}$ (Proposed)	330/330/100 with flippers 490/330/100	$\mathbf{3.68}$	$\mathbf{\approx 4.00}$	1. RGB Camera, 2. ToF Camera, 3. IMU, 4. 2D Lidar, 5. Head Lights, 6. Air pumps, 7. TinyPICO, Autonomous: Nvidia Jetson Orin NX	Pneumatic	Yes 2 motors
4	510 PackBot [ 14 ]	686/406/178 with flippers 889/521/178	$10.9$	$\approx 4.00$	All	Any	Yes 2 motors
5	Dr. Robot [ 8 ]	640/655/176 with flippers 820/655/176	$21.5$	$4.00$	All	Any	Yes 2 motors

Table 1: Comparison of miniUG

V_{2}

with Existing Mobile Platforms

Figure 6: Illustration of the miniUGV 2 navigating three distinct scenarios. From left to right: the first image demonstrates its navigation capabilities within a cluttered indoor environment; the second image showcases the miniUGV 2 traversing a sharp border edge which is a particularly challenging task for a robot of this form factor; and the third image highlights its ability to navigate uneven terrain composed of gravel and concrete structures.

Figure 7: Illustrates the camera, LiDAR views and the position of the articulated arms before initiating a pick operation.

Figure 8: Illustration of the initial state of UAV deploying the miniUGV 2 : This figure captures the snapshot just before the deployment process begins. The left panel shows the point cloud generated from the UAV’s downward-facing RGB-D camera. The coordinate frame at the center of this image represents the camera frame. Yellow points correspond to visual features extracted from the RGB image and projected into 3D space. These same visual features are highlighted in the other two images. The top-right panel displays the raw RGB camera view, while the bottom-right panel shows the colorized depth image, where blue points indicate areas farther from the camera and green points indicate areas closer to it.

Figure 9: This Figure illustrates the scenario after landing when the 3D mapping process is resumed. Note that the map update is paused during the deployment. The posistion of miniUGv 2 is determined by the point cloud processing and shown with green sphare. The colored point cloud is shown to verifiy that the position estimation is accurate and it is inside the point cloud cluster of the miniUGV 2 . The opening of the hidden area is also clearly visible in the 3D map. The black square shown the miniUGV 2 in both the RGB image and the 3D point cloud map for reference.

Robotics

Robotics studies the design and control of machines that perceive, move, and act in physical environments.

See all 751 papers

Seeing Where to Deploy: Metric RGB-Based Traversability Analysis for Aerial-to-Ground Hidden Space Inspection
Mar 15, 2026Seoyoung Lee, Shaekh Mohammad Shithil, Durgakant Pushp +2Robotics Robotic Inspection
The University of Texas at Austin, USA. · Indiana University, Bloomington, USA.
Towards Reliable Aerial Ground Vehicle Collaboration: An Integrated Planning and Autonomy Framework for Field Deployment
Jul 7, 2026Md Safwan Mondal, Luca Russo, James D. Humann +2Vehicle Routing Multi-Robot Coordination
Department of Mechanical and Industrial Engineering, University of Illinois Chicago, IL, 60607 USA · DEVCOM Army Research Laboratory, Minneapolis, Minnesota, USA · DEVCOM Army Research Laboratory, Aberdeen Proving Grounds, Aberdeen, MD 21005 USA
A Tilt-Rotor UAV with a Gripper for Stable Contact-Based Tasks via Environmental Anchoring
Aug 3, 2026Joshua Taylor, Nursultan Imanberdiyev, Wei-Yun Yau +2Aerial Robotics Contact-Rich Robotic Manipulation
National University of Singapore (NUS), Singapore. · Institute for Infocomm Research (I2R), A*STAR, Singapore.

Mar 15, 2026J/K move · Enter open · S save

Seeing Where to Deploy: Metric RGB-Based Traversability Analysis for Aerial-to-Ground Hidden Space Inspection

Seoyoung Lee, Shaekh Mohammad Shithil, Durgakant Pushp, Lantao Liu+1

The University of Texas at Austin, USA. · Indiana University, Bloomington, USA.

Robotics Robotic Inspection Aerial Robotics 3D Reconstruction

Open paper AskPDF arXiv

Inspection of confined infrastructure such as culverts often requires accessing hidden spaces whose entrances are reachable primarily from elevated viewpoints. Aerial-ground cooperation enables a UAV to deploy a compact UGV for interior exploration, but selecting a suitable deployment region from aerial observations requires metric terrain reasoning involving scale ambiguity, reconstruction uncertainty, and terrain semantics. We present a metric RGB-based geometric-semantic reconstruction and traversability analysis framework for aerial-to-ground hidden space inspection. A feed-forward multi-view RGB reconstruction backbone produces dense geometry, while temporally consistent semantic segmentation yields a 3D semantic map. To enable deployment-relevant measurements without requiring LiDAR-based dense mapping, we introduce an embodied motion prior that recovers metric scale by aligning predicted camera motion with onboard platform egomotion. From the metrically grounded reconstruction, we construct a confidence-aware geometric-semantic traversability map and evaluate candidate deployment zones under reachability constraints. Experiments on a tethered UAV-UGV platform demonstrate reliable deployment-zone identification in hidden space scenarios.

Abstract

Robotics

Explore similar work

Seeing Where to Deploy: Metric RGB-Based Traversability Analysis for Aerial-to-Ground Hidden Space Inspection

Towards Reliable Aerial Ground Vehicle Collaboration: An Integrated Planning and Autonomy Framework for Field Deployment

A Tilt-Rotor UAV with a Gripper for Stable Contact-Based Tasks via Environmental Anchoring