Adaptive Robotics & Technology Lab

ART Lab

Research

The ART Lab conducts research at the intersection of robotics, automation, and intelligent systems, with an emphasis on scalable, resilient, and field-deployable solutions. Our work spans robot mechanism design, multi-robot systems, and data-driven autonomy, integrating hardware, algorithms, and real-world experimentation. Through close collaboration with multidisciplinary research teams, the ART Lab advances robotic technologies that address practical challenges in manufacturing, agriculture, and unstructured environments. Below, we highlight representative robots and ongoing projects from the ART Lab.

Decision making and collaborative task performance are fundamental engineering challenges in artificial swarm systems. This line of research focuses on the development of distributed algorithms for consensus-based decision making, task allocation, and coordinated collective behaviors, enabling robotic swarms to operate robustly without centralized control. In parallel, it investigates the design and integration of effective, scalable, and reliable wireless communication frameworks that support information sharing, coordination, and resilience under dynamic and uncertain operating conditions. Together, these efforts aim to establish the algorithmic and networking foundations necessary for deploying large-scale, autonomous robotic swarms in real-world environments.

PUBLICATIONS:

BEING UPDATED.

PUBLICATIONS:

  • Gad S, Bagavathiannan M, Bhandari M, Cason J, Hardin R, Landivar J, Lee K. Low-Cost, Compact Mobile Robot for Autonomous Soil Monitoring in Crop Fields. In2025 22nd International Conference on Ubiquitous Robots (UR) 2025 Jun 30 (pp. 438-443). IEEE.
  • Vemula N, Um D, Bhandari M, Lee K. Hardware Prototype and System Apparatus of an Autonomous Robotic Harvesting Cell. In2025 22nd International Conference on Ubiquitous Robots (UR) 2025 Jun 30 (pp. 413-418). IEEE.
  • Wei Y, Johnson J, Bagavathiannan M, Lee K. An Integrated System for Multirobot-Based Autonomous Scouting of Agricultural Crop Fields. Available at SSRN 5183985.
  • Pal P, Landivar-Bowles J, Landivar-Scott J, Duffield N, Nowka K, Jung J, Chang A, Lee K, Zhao L, Bhandari M. Unmanned aerial system and machine learning driven Digital-Twin framework for in-season cotton growth forecasting. Computers and Electronics in Agriculture. 2025 Jan 1;228:109589.
  • Nethala P, Um D, Vemula N, Montero OF, Lee K, Bhandari M. Techniques for Canopy to Organ Level Plant Feature Extraction via Remote and Proximal Sensing: A Survey and Experiments. Remote Sensing. 2024 Nov 22;16(23):4370.
  • Stewart DG, Mendes ED, Lee K, Blum ME, Tedeschi LO, Webb SL. Comparison of GPS Collars and Solar-Powered GPS Ear Tags for Animal Movement Studies. Smart Agricultural Technology. 2025 Aug 1;11:101021.

Passive Wheel-Leg Mechanism

In swarm robotics, it is desirable for each robot to be as small and mechanically simple as possible. However, reducing robot size often comes at the cost of limited locomotion capability—small wheels struggle on uneven or rough terrain. Legged robots offer greater agility and adaptability, but they are difficult to control and mechanically complex to realize in a compact form factor. Passive wheel–leg transformable mechanisms (see a picture on the left) bridge this gap by combining the advantages of wheeled and legged locomotion without introducing additional structural or control complexity. Building on this design and its proof-of-concept prototype, we are currently scaling up this mechanism to develop fully equipped unmanned ground vehicles (UGVs), referred to as α-WaLTR, with versatile locomotion capabilities. To learn more about this project, please see our previous post. A two-wheel version of the robot with water resistant hardware features (named ARMoR) was also developed for amphibious locomotion on water surface and diverse terrains. 

RELATED PUBLICATIONS

Associated IP/Patents:

  • Transformable Wheel and Leg Assembly, US12103331B2, US Utility Patent 2024

CLAW

Cycloidal Leg-Augmented Wheel (CLAW) integrates three leg segments with a wheel using a specialized passive bar mechanism inspired by a cycloidal rotor design. These legs extend outward to reach heights from the front and retract back within the wheel boundary as they approach the ground. This unique mechanism enables the legs to generate cycloidal trajectories as the wheel rotates without additional actuators, thereby preserving the operation and control simplicity of conventional wheeled robots while significantly improving climbing abilities.

RELATED PUBLICATIONS:

Origami offers a novel source of inspiration for the design of semi-soft robotic mechanisms by introducing a fundamentally different approach to structural compliance. Rather than relying on material deformation, as is common in many soft robots, origami—particularly action origami that exhibits intrinsic kinematic motion—achieves a unique coexistence of flexibility and rigidity through geometry alone.

Figure. hand-folded origami twisted tower.

One particular origami design of our interest is called Twisted Tower by Mihoko Tachibana (See the picture above). TWISTER was our first 3D-printed origami twisted-tower structure that faithfully preserves the exact kinematic behavior of this original origami design. To enable physical realization, we developed multiple 3D modeling strategies that significantly improve structural durability when fabricated using both flexible and rigid materials. Building on this design, the TWISTER Hand employed three such 3D-printed origami structures as gripper fingers, enabling versatile and adaptive object manipulation.

 

RELATED PUBLICATIONS:

 

Figure. The e-Cube games for cognitive assessment and training.

This project develops and evaluates a tangible block-based game platform for automated, play-based assessment of cognitive skills. The original system consisted of a set of highly instrumented blocks equipped with motion sensing and wireless communication capabilities to capture players’ manipulation patterns, speed, and behavioral metrics. To reduce technical complexity and deployment cost, we have recently transitioned this platform to a vision-based approach that uses 3D-printed one-inch cubic blocks and an overhead camera. We are currently conducting a human-subject study to evaluate the effectiveness of the vision-based tangible games for cognitive assessment.

A variety of interesting games, incorporating music and auditory feedback into the game design have been also developed. The above video highlights these MusicBlock games (David Miranda, MS 2018)

PUBLICATIONS: