Live Intelligence Engine

SomaticAI Engine

Watch raw IMU sensor data flow through the Temporal Convolutional Network in real time — from physical motion to labelled and annotated intelligence.

Raw IMU Data

IMU Sensor Stream — 100Hz

1{ax:-1.535, ay:-8.019, az:-0.737, gx:-0.115, gy:-0.214, gz:-0.197}

2{ax:-1.525, ay:-8.009, az:-0.727, gx:-0.114, gy:-0.213, gz:-0.196}

3{ax:-1.515, ay:-7.999, az:-0.717, gx:-0.113, gy:-0.212, gz:-0.195}

4{ax:-1.505, ay:-7.989, az:-0.707, gx:-0.112, gy:-0.211, gz:-0.194}

5{ax:-1.495, ay:-7.979, az:-0.697, gx:-0.111, gy:-0.210, gz:-0.193}

6{ax:-1.485, ay:-7.969, az:-0.687, gx:-0.110, gy:-0.209, gz:-0.192}

↑ live packets @ 1293ms latency

→

SomaticAI
Engine

Temporal Convolutional Network (TCN)

Input

TCN Layer 1

TCN Layer 2

TCN Layer 3

Output

Receptive field: 512 timesteps · Dilation: 1, 2, 4, 8

→

Labelled Output

Current Activity

sweeping

Confidence94.0%

sweeping

mopping

lifting

carrying

Annotated Output

joint_angle_shoulder

96.8

joint_angle_elbow

66.3

joint_angle_hip

36.3

force_vector_z

69.9

ergonomic_risk

9.7

temporal_segment

54.7

Output format: JSON · 6 annotation fields · 100Hz resolution

Architecture Overview

Input Layer

· 6-axis IMU (accel + gyro)
· 100Hz sampling rate
· Sliding window: 512 samples
· Normalisation + detrending

TCN Core

· 4 dilated causal conv layers
· Dilation factors: 1, 2, 4, 8
· Residual connections
· Dropout: 0.2 (training only)

Output Head

· Softmax classification (8 classes)
· Confidence score per label
· Joint angle estimation
· Force vector annotation

Trade Secret Protection

The SomaticAI Engine model weights, training pipeline, and annotation methodology are protected as trade secrets. The architecture shown above is a simplified representation for demonstration purposes.