Wearable Humanoid Robot Control for Real-Time Motion Data Capture
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot training systems are bulky, expensive, and require direct control and constant access to physical robots, making data collection difficult, unsafe, or inefficient, and they struggle with limited training efficiency and high hardware dependency.
Innovation Solution
A portable, lightweight wearable data collection apparatus that allows operators to control robots and collect training data in simulated or robot-free environments, using the same sensor technology as the robot to ensure accurate data capture.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional robot training systems are used, then data collection can be performed, but the systems are bulky, expensive, and require direct control and constant access to physical robots
Solution Approach 1:
The patent uses motion capture technology to create digital copies of human movements and robot movements in virtual environments. Operators control virtual robot avatars that replicate physical robot actions, allowing data collection without direct physical robot access. This copying approach eliminates the need for bulky physical training setups while maintaining data collection reliability
Solution Approach 2:
The patent replaces mechanical robot training systems with software-based virtual training environments. Instead of requiring physical robots and complex mechanical training apparatus, the system uses computer-generated virtual robots and motion capture data to achieve the same training objectives, significantly reducing hardware dependency and system bulk
2Measurement precision
If physical robot access is required for data collection, then accurate robot motion data can be captured, but the process becomes unsafe and inefficient
Solution Approach 1:
The patent creates virtual replicas of physical robots and their motion patterns in simulated environments. Motion capture sensors record actual robot movements, which are then copied into virtual robot avatars. This allows operators to collect and analyze robot motion data safely in virtual spaces without requiring constant physical robot access, maintaining measurement precision while improving safety
Solution Approach 2:
The patent introduces motion capture technology and virtual simulation as intermediary layers between operators and physical robots. Instead of directly manipulating physical robots for training, operators interact with virtual representations through the intermediary software system, which accurately translates and preserves the original motion data while eliminating safety risks
3Productivity
If traditional training systems are used, then robot training can be performed, but training efficiency is limited and hardware costs are high
Solution Approach 1:
The patent uses virtual copying of robot training scenarios to enable unlimited repetitions without wearing physical robots or consuming physical resources. Training data can be copied and reused indefinitely in virtual environments, dramatically improving training efficiency while eliminating hardware wear and reducing costs
Solution Approach 2:
The patent creates a universal virtual training platform that can train multiple robot types and configurations using the same software infrastructure. A single motion capture setup and virtual environment can accommodate different robot designs, making the system highly versatile and eliminating the need for dedicated hardware for each training scenario
Data Source
AI summary
The humanoid robot data collection system may include a wearable data collection apparatus, a computer, and a humanoid robot in data communication via a network. The wearable data collection apparatus worn by a human operator to control a robot for the purpose of collecting robot data. The wearable data collection apparatus generates pilot movement data based on various sensors contain in the wearable data collection apparatus, where the pilot movement data is used to generate robot control data and sends this control data to the robot in real-time. The wearable data collection apparatus includes a base frame, articulated arms extending from the base frame, sensors coupled to articulated arms, gloves with sensors coupled to the articulated arms, and a piloting control system operatively connected to the plurality of sensors.