Robot Software Platform
PLEM
ROS 2-Based AI Robot Control Platform
Everything the industry needs, built on top of ROS 2's powerful ecosystem. An all-in-one robot platform integrating real-time control, AI libraries, and Docker deployment
Drivers, RTOS, Communication — never build from scratch again.
From motor control to AI pipelines, PLEM handles the control infrastructure. You focus on the robot application.
Measured Performance
Verified on real hardware — based on KOTCA certified data
Robot Software Platform
PLEM
Hover a layer to explore the architecture
Robot Software Platform
PLEM
Hover a layer to explore the architecture
PLEM Software Stack
From LLM to motor control — the full architecture of WIM's real-time control domain
LLM Layer(Claude / GPT)
High-Level Policy
VLA / Motion Planning
Sensors
Sensor Input (state observation)
SoC (Jetson Orin AGX)
Middleware— ROS 2 / Communication Layer (EtherCAT, CAN)
RL-Trained Motor Controller
— WIM Proprietary
Joint states, sensor feedback
Torque / Velocity / Position commands
Safety Control — safety fallback only, activates on anomaly
Deterministic RT Kernel
SoC (Jetson Orin AGX) — Everything runs inside.
Motor Commands via EtherCAT / CAN
Sensors
SoC (Jetson Orin AGX)
Middleware— ROS 2 / Communication Layer (EtherCAT, CAN)
RL-Trained Motor Controller
— WIM Proprietary
Joint states, sensor feedback
Torque / Velocity / Position commands
Safety Control — safety fallback only, activates on anomaly
Deterministic RT Kernel
SoC (Jetson Orin AGX) — Everything runs inside.
Why LLM Cannot Replace This
LLM
WIM RL Controller
Cycle
500–2000ms
1ms
Output
Text / Code
Motor torque
Adapt
Re-prompt
Re-train
Safety
None
Safety fallback
Core Technologies in PLEM
All technologies integrated into a single software stack
Friction-Compensated Motor Control
Friction compensation and acceleration feedforward for improved position tracking accuracy.
Position RMSE -38.9%Read more →WIM Real-Time Kernel
PREEMPT_RT kernel optimized for Jetson Orin. CPU isolation ensures control loop determinism.
Jitter 318µs → 15µs (21× reduction)Read more →TensorRT AI Vision Pipeline
Edge AI inference via TensorRT + DeepStream. KOTCA-certified accuracy.
275 TOPS · mAP 99.1% · Jetson edgeRead more →WIM Precision Calibration
Precise robot coordinate calibration via Touch Probe + Kabsch algorithm.
Sub-0.5mm precisionRead more →Ultra-Low Latency RT Monitoring
Lock-free data transfer between RT control loop and monitoring threads.
0.3µs avg · 0.7µs P99Read more →Official Certifications
Technical credibility validated by domestic and international accreditation bodies
W-RC GPU Controller
Cycle: 0.0232ms / Jitter: 0.0042ms
W-Ecobot Vision
PP/PE AP 99%, 52fps
W-Board Embedded Controller
Latency 25µs, Power 25W
W-Ecobot Robot Performance
Payload, Repeatability, Speed, Picking
W-Module Performance Evaluation
3 performance categories officially evaluated
Quality, Environmental & Safety Management
International management system certifications
Frequently Asked Questions
What is PLEM's relationship with ROS 2?+
PLEM is a higher-level platform built on top of ROS 2. It leverages the ROS 2 ecosystem while providing additional industrial real-time control, AI libraries, and Docker deployment.
How is the real-time kernel applied?+
PLEM provides a PREEMPT_RT kernel for NVIDIA Jetson Orin. Combined with CPU isolation, it reduces jitter by 21× compared to the standard kernel (318µs → 15µs). Based on KOTCA KST-25-236 certified data.
Do I need to configure TensorRT optimization manually?+
No. PLEM includes a built-in pipeline that automatically converts PyTorch models to TensorRT FP16. CUDA environment setup and DeepStream integration are also included in the package.
Can I use PLEM without W-RC?+
PLEM is optimized for W-RC, but it can be used in any ROS 2 compatible environment. See PLEM Docs for details.
