Explainable Neural Inverse Kinematics for Obstacle-Aware Robotic Manipulation: A Comparative Analysis of IKNet Variants
The study applies explainable neural inverse kinematics to obstacle-aware manipulation on the ROBOTIS OpenManipulator-X.
Evidence notes
- The workflow combines Shapley-value attribution with physics-based obstacle-avoidance evaluation.
- The authors train lightweight IKNet variants on synthetic pose-joint data for the OpenManipulator-X platform.
Company context
Develops actuator systems, control platforms, and modular robotics kits, best known for its DYNAMIXEL smart servos used across research, education, and commercial robotics. The company also builds complete robotic platforms including humanoid and mobile robots, primarily as development and ecosystem tools.