One-Leg Exoskeleton Control Using Vision-Based Terrain Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing orthotic devices, particularly exoskeletons for individuals with asymmetric gait impairments due to conditions like hemiplegia or hemiparesis, lack effective electronic control systems that address the unique challenges of single-leg mobility, including terrain irregularities and user intention inference, leading to safety and comfort issues during extended walking.
Innovation Solution
A system for controlling a one-leg exoskeleton using a computer vision unit, inertial sensors, and actuators to infer user gait phase and terrain hazards, enabling adaptive responses through a hierarchical control scheme that integrates machine learning algorithms to adjust actuator movements for safe and comfortable locomotion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If electronic control systems are added to orthotic devices, then gait assistance capability is improved, but device complexity increases
Solution Approach 1:
The control system is divided into multiple independent modules: computer vision unit for terrain detection, inertial sensors for gait phase detection, processor unit for intention inference, and actuator control modules. Each module performs a specific function and can be independently developed and maintained, reducing overall system complexity while maintaining high adaptability.
Solution Approach 2:
The system performs preliminary detection of terrain hazards and user movement intentions before executing gait assistance actions. The computer vision unit continuously scans the environment ahead of the user, and the processor unit predicts user intentions based on inferred gait phase, allowing the system to prepare appropriate responses in advance, improving responsiveness without increasing operational complexity.
2Reliability
If terrain hazard detection is implemented in real-time, then safety is improved, but processing time and computational resources increase
Solution Approach 1:
The computer vision unit continuously captures and pre-processes terrain information in advance before hazards become immediate threats. By maintaining a ready buffer of processed terrain data and hazard classifications, the system can quickly respond to new hazards without requiring intensive real-time computation during critical moments, thus improving safety response time.
Solution Approach 2:
The system focuses computational resources on detecting and analyzing only the most critical terrain hazards and the immediate walking path ahead of the user. Rather than processing entire environmental scenes uniformly, the computer vision unit prioritizes analysis of regions with higher risk potential, reducing overall processing time while maintaining high safety standards for critical areas.
3Measurement precision
If user intention inference is added to improve responsiveness, then gait assistance accuracy is improved, but system complexity increases
Solution Approach 1:
The processor unit continuously monitors inertial sensor data from IMUs and pressure sensors to infer user gait phase and movement intentions. This inferred information feeds back to the actuator control modules, which adjust assistance timing and magnitude accordingly. The feedback loop enables accurate, adaptive gait assistance by aligning actuator activation with the user's natural gait cycle and intended movements, improving precision without requiring complex mechanical systems.
Data Source
Figure 1
Figure 2A~2D
Figure 3A~3C
AI summary
A system (100) for controlling a one-leg exoskeleton (200) for walking assistance of a user is disclosed. The exoskeleton (200) comprises articulated units movable by a first actuator (210) and a second actuator (220). The system (100) comprises a plurality of inertial mass units (302,304,306) and pressure sensor (308) to collect gait information. The system (100) comprises a computer vision unit (110) to collect environment information associated to a user path and to detect a terrain hazard (122) and its geometrical attributes. A user gait phase (126) is inferred via a processor unit (120) comprised in the system (100) using additional gait information collected from the actuators (210,220). A response is produced with selective instructions for the actuators (210, 220), based, at least, on the geometrical attributes of the terrain hazard (122), a user profile (120) with user limitations and a user intention (128) obtainable from an interface (270, 260).