PREFAIL: Predictive Feasibility From Accumulated Introspective Logs
Author: Saurabh Mahesh Gaglani Research Area:Task & Motion Planning · AI & LLM Robotics Summary:A predictive feasibility framework using myCobot 280 as a low-cost yet challenging research platform to identify physically infeasible robot manipulation tasks before motion planning. Product used:myCobot 280 Pi Background:LLM-generated robot tasks may appear valid but fail due to physical constraints such as workspace, joint limits, grasp geometry, and payload. Key Challenges:● Predicting task feasibility before expensive...
Find the Fruit: Designing a Zero-Shot Sim2Real Deep RL Planner for Occlusion Aware Plant Manipulation
Author：Nitesh Subedi , Hsin-Jung Yang, Devesh K. Jha ,Soumik Sarkar Research Area: Agricultural Robotics · Reinforcement Learning Summary: A zero-shot Sim-to-Real robotic manipulation system using Elephant Robotics MyBuddy 280 with RGB-D vision and PPO to actively move plant structures and reveal occluded fruits for autonomous harvesting. Product used: myBuddy 280 Pi Background: Occluded fruits are difficult for harvesting robots to locate, especially in dense and deformable plants. Traditional methods struggle...
Front Hair Styling Robot System Using Path Planning for Root-Centric Strand Adjustment
Author：Soonhyo Kim , Naoaki Kanazawa , Shun Hasegawa , Kento Kawaharazuka and Kei Okada Research Area: Robot Manipulation · Motion Planning Summary: A robotic hair styling system using Elephant Robotics myCobot 280 with a comb end effector to generate root-centric combing trajectories and iteratively adjust hair strands toward a target hairstyle. Product used: myCobot 280 M5 Background: Robotic hair styling requires precise manipulation of hair strands and effective path planning...
High flexibility of heterogeneous tri-robot collaborative handling
Author：ZHANG Shuzhong ，QI Chunyu ，ZHANG Gong，SU Jiahong，QIU Weiqian，RUAN Yuzhen Research Area: Multi-Robot Collaboration · Reinforcement Learning Summary: A PPO-based collaborative handling system using Elephant Robotics myCobot 280 and myPalletizer with an AUBO i5 to enable heterogeneous robots to coordinate object handling with smoother motion and improved trajectory tracking. Product used: myCobot 280 M5 myPalletizer 260 M5Stack Background: Heterogeneous multi-robot collaboration can handle large or heavy objects beyond the capabilities of...
Inverse Kinematics for a 6-Degree-of-Freedom Robot Manipulator Using Comprehensive Gr¨obner Systems
Author：Takumu Okazaki , Akira Terui ，and Masahiko Mikawa Research Area: Inverse Kinematics · Robot Kinematics Summary: A Comprehensive Gröbner System-based inverse kinematics method using Elephant Robotics myCobot 280 to efficiently solve and accelerate repeated inverse kinematics calculations for a 6-DOF robotic arm. Product used: myCobot 280 M5 Background: Efficient inverse kinematics is essential for precise motion control and trajectory planning of 6-DOF robotic arms. This research uses Elephant Robotics’ myCobot...
Multi-Objective Optimization for Inverse Kinematics of Robotic Arm with Joint Configuration
Author：Xibin Zhou, Tanyaradzwa Chinyai, Lizhe Tan Research Area: Inverse Kinematics · Motion Optimization Summary: A multi-objective PSO-based inverse kinematics optimization method using Elephant Robotics myCobot 280 Pi to select efficient joint configurations and reduce unnecessary joint rotation during trajectory execution. Product used: myCobot 280 Raspberry Pi Background: Traditional inverse kinematics can produce multiple solutions and may not provide an energy-efficient joint configuration. This study focuses on optimizing robotic arm motion...
Method for Determining Motion Paths Satisfying Placement Method and Singularity Avoidance in Cooperative Motion Using Multiple Cooperative Robots
Author：Kenta Naka, Daiki Kato, Shota Tsuzioka, Masao Nakagawa, Toshiki Hirogaki, Eiichi Aoyama Research Area: Multi-Robot Coordination · Motion Planning · Singularity Avoidance Summary: A cooperative motion planning method using two Elephant Robotics myCobot 320 M5 robots to select motion paths that reduce joint loads, avoid singularities, and improve motion accuracy. Product used: myCobot 320 M5 Background: Cooperative robots can follow multiple motion paths, but choosing an optimal path that balances...
Enhancement of Control Performance for Degraded Robot Manipulators Using Digital Twin and Proximal Policy Optimization
Author：SU-YOUNG PARK , CHEONGHWA LEE , HYUNGJUNG KIM , AND SUNG-HOON AHN Research Area: Robot Control · Reinforcement Learning · Digital Twin Summary: A digital-twin and PPO-based control framework using Elephant Robotics myCobot 280 Pi to compensate for degraded robot performance and improve real-world positioning accuracy through sim-to-real transfer. Product used: myCobot 280 Raspberry Pi Background: Robot aging and actuator degradation can cause positioning errors, oscillation, and unstable motion, while...