Robot Posture Coaching and Learning for Adaptive Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current robotic systems lack the ability to enhance existing functions and acquire new ones through learning, limiting their entertainment capabilities and user engagement.
Innovation Solution
A robotic device with a drive mechanism, control unit, and mode setting unit that allows for coaching and learning modes, enabling users to instruct and improve the robot's posture and control modes, allowing it to learn and adapt new functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the robot uses fixed control programs for basic functions, then the control system is simple and reliable, but the robot cannot acquire new functions or enhance existing ones
Solution Approach 1:
The control system transitions from static fixed programs to dynamic learning capabilities. The robot can adapt its control modes by learning from coaching data, allowing the system to evolve and acquire new functions while maintaining a relatively simple hardware architecture.
Solution Approach 2:
The robot performs self-learning and self-improvement through the learning unit that processes coaching data. Instead of requiring external reprogramming, the robot autonomously derives new control modes by learning from demonstrated postures and instructions.
2Ease of operation
If the robot operates in autonomous mode only, then the operation is simple, but the user cannot provide coaching or guidance
Solution Approach 1:
The control unit is designed to handle multiple operation modes (autonomous mode and coaching mode) within a single unified system. This allows the robot to switch between autonomous operation and user-guided learning, providing versatility without requiring separate dedicated systems for each mode.
Solution Approach 2:
The operation mode is made dynamic and switchable rather than fixed. The robot can transition between autonomous operation and coaching reception based on user input, allowing flexible interaction while maintaining simple autonomous operation when coaching is not needed.
3Manufacturing precision
If the robot learns complex postures through coaching, then the performance improves, but the learning time and processing load increase
Solution Approach 1:
The coaching unit captures and stores demonstration postures and instructions in advance during the coaching phase. This preliminary action allows the learning unit to process and derive control modes without real-time computation delays during actual execution, reducing learning time while maintaining precision.
Solution Approach 2:
The system focuses on learning key posture parameters and essential control modes rather than every detail of movement. By concentrating on the most important aspects of posture control, the robot achieves high precision while minimizing the computational burden and learning time.
Data Source
AI summary
A mode setting unit sets any one of operation modes in an operation mode group including at least a coaching mode and a learning mode. In the coaching mode, a control unit receives a posture instruction and controls a storage unit to store the posture instruction. In the learning mode, the control unit derives a control mode of a drive mechanism by learning while reflecting, in a posture of the robotic device, the posture instruction received in the coaching mode.