Open · Efficient · Intelligent

Document version: V1.0 Hardware platform: INEXBOT industrial robot controller (RK3588 platform) Application fields: dual-arm collaboration, human-robot collaboration, robot algorithm verification, multi-machine coordinated control, advanced control theory research, industrial manufacturing, medical care, commercial services, logistics and warehousing, education and scientific research

---

1. Company Profile

INEXBOT Technology is a technology company focused on the R&D of robot motion control systems, with self-developed advanced control algorithms and rich industry experience, committed to promoting the application and implementation of wheeled humanoid robot technology in industrial and service fields. The INEXBOT motion control system can achieve precise control of key parts such as dual arms, waist, and neck, endowing robots with excellent flexibility and adaptability.

---

2. Product Overview

The INEXBOT wheeled humanoid robot motion control system is designed for complex scenarios, supporting multiple configuration types to meet different application needs. The platform is based on the ABOX-6116 industrial controller (Rockchip RK3588 + Debian 11 / Ubuntu 20.04) as the hardware foundation, standardly equipped with dual robot control engines (total capacity of 14 axes or 12 axes, expandable up to 32 axes), opening the full-chain NexDroid OpenAPI interfaces, supporting users' complete autonomous algorithm development from the joint control layer to the trajectory planning layer. The system adopts a modular design philosophy with strong compatibility, and can seamlessly interface with third-party wheel bases, dexterous hands, and vision systems.

The platform is oriented toward universities, research institutes, and enterprise laboratories, providing a highly real-time, highly open hardware foundation and software ecosystem for scientific research scenarios such as dual-arm coordinated motion control, advanced dynamics algorithms, and intelligent control strategies, as well as practical implementation in industrial and service fields.

---

3. Product Features

3.1 Dual-Arm Collaborative Control System

• Supports two dual-arm configuration schemes:
  • Dual six-axis collaborative robots (2×6 axes) configuration
  • Dual seven-axis collaborative robots (2×7 axes) configuration
• Seven-axis configuration: the wrist supports SRS structure (spherical joint - revolute joint - spherical joint) or cross-cross structure, flexibly selectable according to operational requirements
• Equipped with high-precision trajectory planning and collision detection functions
• Supports dual-arm collaborative operation modes, enabling:
  • Symmetric synchronized motion
  • Asymmetric collaborative operation
  • Object handover and transfer
  • Force-position hybrid control

3.2 Waist Control System

• Provides multiple configuration options:
  • 1 axis (rotation)
  • 3 axes (X/Y/Z translation)
  • 4 axes (4 serial joints)
  • 6 axes (3 translation + 3 rotation)
• Supports dynamic balance compensation
• Has load-adaptive capability

3.3 Neck Control System

• Single-axis rotation design
• Expandable to multi-DOF configurations
• Supports vision following function

3.4 Chassis Control System

Item	Description
Chassis structure	Dual differential wheels + auxiliary caster wheels
Communication protocol	CAN bus, directly controls the differential wheels

---

4. Application Scenarios

The INEXBOT wheeled humanoid robot motion control system is widely used in the following fields:

4.1 Industrial Manufacturing

• Automotive assembly lines: can complete precision operations requiring dual-arm collaboration such as door installation and dashboard assembly
• Electronics manufacturing: suitable for fine operations such as circuit board assembly and component welding
• Machining: can perform auxiliary work such as tool change and part handling

4.2 Medical Care

• Surgical assistance: high-precision dual arms can assist doctors in precision operations
• Rehabilitation training: can provide stable support and assistance
• Medicine dispensing: precise operations in sterile environments

4.3 Commercial Services

• Catering service: completes tasks such as tableware placement and beverage preparation
• Hotel reception: provides luggage handling and guidance services
• Retail display: product display and interactive demonstrations

4.4 Logistics and Warehousing

• Shelf restocking: dual arms collaboratively complete cargo grasping and placement
• Package sorting: cooperates with vision systems for efficient sorting
• Heavy material handling: the waist support system can bear larger loads

4.5 Education and Scientific Research

• As a teaching and research platform, the robot can be used for experiments and development in fields such as artificial intelligence and robotics

4.6 Dual-Arm Assembly

• Master arm holds the workpiece, slave arm holds the tool
• Compliant assembly strategy based on torque sensing
• Force-position hybrid control

4.7 Dual-Arm Handling

• Collaborative grasping of large-size/irregular workpieces
• Master-slave posture-constrained motion
• Dual-arm load distribution optimization

4.8 Dual-Arm Polishing/Buffing

• Master arm positions the workpiece, slave arm performs polishing
• Constant force tracking control
• Trajectory coordination compensation

4.9 Algorithm Verification Platform

• Rapid deployment of new control laws (MPC/RL/adaptive)
• Dynamics parameter identification
• Dual-arm impedance/admittance control
• Human-robot physical interaction (pHRI) algorithms

---

5. Product Advantages

5.1 High-Performance Processing Platform

• Equipped with Rockchip RK3588 processor (8 cores, 4×Cortex-A76 + 4×Cortex-A55, up to 2.4GHz, built-in 6 TOPS NPU, expandable to 20 TOPS), see Section 7: Hardware Specifications
• Onboard LPDDR5 8GB memory + 64GB eMMC storage
• Fanless aluminum-magnesium alloy body, operating temperature -20°C ~ 70°C

5.2 Real-Time Motion Control

• Uses EtherCAT communication, high speed and high precision, slave station synchronization jitter <20μs
• Control cycle up to 1ms (position loop/speed loop), minimum 250μs
• Self-developed motion control algorithm, supports dual-arm collaboration

5.3 Flexible Configuration

• Waist supports multiple axis configurations (1-6 axes) to suit different task needs
• Dual arms compatible with six-axis and seven-axis robots, enhancing operational diversity
• Dual arms with independent configurations, supporting serial multi-joint, SCARA, and custom configurations
• DH parameters open and configurable, supporting online modification

5.4 Excellent Openness and Compatibility

Communication protocols:

• EtherCAT (master/slave)
• PROFINET RT/IRT
• Ethernet/IP
• Modbus TCP/RTU
• CANopen
• OPC-UA

Physical interfaces are detailed in Section 7.2, and software protocols in Section 8.3.

Third-party device access support:

• Wheeled chassis (via CAN bus)
• Dexterous hands (supporting Modbus/TCP, Modbus/RTU protocols)
• Vision systems (EtherNet)

Standard API interfaces provided:

• ROS/ROS2 support
• Python, C#, C++ SDK

---

6. Motion Control System Specifications

6.1 Basic Parameters

Item	Specification
Dual-arm axes	Dual arms total 14 axes (left arm 7 axes + right arm 7 axes), or dual arms total 12 axes (left arm 6 axes + right arm 6 axes)
Maximum axes	32 axes (total system capacity, including external axes)
Control cycle	1ms (position loop/speed loop), minimum 250μs
EtherCAT synchronization	Distributed clock (DC), slave station synchronization jitter 20μs
Multi-axis synchronization error	< 50μs
Interpolation methods	PTP / Linear / Arc / Spline curve / FLYBY
Coordinate systems	Joint coordinate system, robot coordinate system, tool coordinate system, user coordinate system
Robot configuration	Dual arms with independent configurations, supporting serial multi-joint, SCARA, custom configurations; seven-axis wrist supports SRS / cross-cross structures
DH parameters	Open and configurable, supporting online modification
Repeat positioning accuracy	±0.005mm
Maximum motion speed	2m/s
Maximum acceleration	5m/s²
Force control resolution	0.1N
Torque control accuracy	±0.5% FS

6.2 Dual-Arm Collaboration Modes

Feature	Description
Collaboration modes	Dual robot coordination (master-slave/loose coupling/tight coupling)
Synchronization accuracy	Fully synchronized start/stop of both robots, based on EtherCAT distributed clock
Interference zones	Supports dual-arm interference area setting and collision protection
Coordinate system association	Supports defining slave arm workspace under master arm base coordinate system
Coordination instructions	Dual-robot point-to-point (CoordPTP), dual-robot linear (CoordLine), dual-robot arc
Independent operation	Each robot's parameters/programs/variables are independent, and synchronized start/stop is also supported

6.3 Control Modes

Mode	Support Level	Description
Joint position mode	Position loop	Given joint angle target, the controller plans the trajectory and executes it
Joint velocity mode	Speed loop	Directly issues joint velocity commands, open-loop or closed-loop both supported
Joint torque/current mode	Current loop	Directly issues joint torque commands (requires matching servo support)

---

7. Hardware Specifications

7.1 Controller

Item	Parameter
Model	ABOX-6116 industrial controller
CPU	Rockchip RK3588 8-core (4×Cortex-A76 + 4×Cortex-A55, up to 2.4GHz)
GPU	Mali-G610 MC4
NPU	6 TOPS (INT8); optional M.2 RK1820 compute card expansion to 20 TOPS
Memory	Onboard LPDDR5 8GB
Storage	Onboard 64GB eMMC; optional M.2 2280 SATA3 SSD
Network	5× Gigabit Ethernet (4× Intel i210 + 1× RTL8211F)
Wi-Fi / BT	Onboard 2.4G/5G dual-band Wi-Fi 6 + Bluetooth 5.0
4G LTE	1× Mini-PCIe, supports LTE 4G module (optional)
Operating system	Ubuntu + INEXBOT control system
Power supply	DC 12~24V ±10%, power consumption ≤20W, overcurrent/overvoltage/reverse connection protection
Cooling method	Fanless design, aluminum-magnesium alloy shell + aluminum profile heat dissipation
Dimensions	170mm × 112mm × 48mm (excluding bracket)
Weight	1.5kg
Mounting method	Wall mount / DIN-Rail guide rail mounting
Operating temperature	-20°C ~ 70°C

7.2 Communication and Control Interfaces

Interface	Quantity	Description
Gigabit Ethernet	5 ports	4× Intel i210 + 1× RTL8211F, 1 port used as EtherCAT master station, supporting distributed clock (DC) synchronization
USB 3.0	2 ports	1 port multiplexed as OTG, single port supports 5V@2A
USB 2.0	1 port	Peripheral access
HDMI	1 port	Supports up to 8K (7680×4320) output
RS232	2 channels	Isolated, 3.5mm Phoenix terminals
RS485	2 channels	Isolated, 3.5mm Phoenix terminals, supports Modbus RTU
Isolated DI	8 channels	Dry/wet contact auto-switching, isolation voltage 2500Vrms
Isolated DO	8 channels	NPN/PNP switchable (DIP switch), isolation voltage 2500Vrms
PWM output	4 channels	Light source control, supports external hardware trigger (DC 5~24V), single channel max 1A (optional)
Micro SIM	1 port	4G LTE module SIM card

7.3 Servo Drive System (Selection Recommendation)

Solution	Recommended Model	Description
EtherCAT servo	EtherCAT joint modules compliant with CIA402 standard	-
Encoder type	Supports absolute encoders	—

7.4 Teach Pendant (Optional)

Item	Parameter
Model	T30 robot-dedicated teach pendant
Screen	8-inch TFT full touchscreen, 1024×600
Operating system	Linux + QT
Functions	Teach programming, parameter configuration, status monitoring, program debugging

7.5 Environmental Requirements

Item	Requirement
Power supply	DC 12~24V ±10% (≤20W, 12V 40W adapter recommended)
Operating temperature	-20°C ~ 70°C
Servo rated current	Selected according to the manipulator arm specifications
Host computer	Windows / Linux workstation, Gigabit Ethernet port

---

8. Software Platform Specifications

8.1 System Architecture

8.2 Secondary Development Interfaces (NexDroid OpenAPI)

Dual-arm control:

Interface Category	Content	Development Language
Joint control	Joint position/velocity/torque command issuance	C#, Python, TCP Socket
Kinematics interfaces	Forward kinematics (FK) input/output access	C#, Python, TCP Socket
Trajectory planning	Custom trajectory generation and download path	C#, Python, TCP Socket
Admittance control	Simultaneously issue target position + stiffness parameters (K) + damping parameters (D) for compliant interaction	C#, Python, TCP Socket
Status feedback	Real-time joint position, velocity, torque, IO status reading	C#, Python, TCP Socket
Variable system	Global/local variable read/write	C#, Python, TCP Socket
Servo control	Servo power on/off, alarm clearing	C#, Python, TCP Socket
DH parameters	Online modification of robot DH parameters	Teach pendant / configuration file
Configuration management	Controller configuration import/export	File system

Chassis control:

Category	Interface	Description
Chassis management	Enable	Enable the chassis and enter the controllable state
	Emergency stop	Immediately stop chassis motion
	Clear error	Clear chassis fault/alarm states
	Control mode switching	Supports remote control mode, navigation mode, idle mode
Direct control	Velocity/angular velocity issuance	Issue linear velocity and angular velocity via protocol to control chassis forward, backward, turning, and in-place rotation
Navigation control	Path tracking	Issue path point sequences for tracking, supporting forward and reverse paths
	Path replacement	Replace current path points during operation
	Replanning	Trigger path replanning
	In-place rotation	Rotate in place to a specified heading in navigation mode
	Stop navigation	Abort the current navigation task
Status and configuration	Current status query	Query the chassis real-time status (position, velocity, mode, etc.)
	Set odometry parameters	Configure odometry resolution, wheelbase, and other parameters
	Set heading angle	Set/correct the chassis current heading angle

8.3 Communication Protocols

Protocol	Purpose	Features
Port 7000 protocol	Core control protocol (JSON over TCP)	Query/issue joint coordinates, variables, IO, supports single-point and continuous trajectories
TCP custom communication	Host computer ↔ controller	Server/Client mode, supports 9 process numbers concurrently
Modbus TCP/RTU	PLC/sensor integration	Flexible master/slave configuration
OPC-UA	Data acquisition/monitoring systems	Standardized industrial interconnection
EtherNet/IP	Allen-Bradley PLC	CIP protocol support
FinsTCP	Omron PLC	Command/response format

Physical layer communication performance: EtherCAT minimum cycle 100μs, PROFINET IRT minimum cycle 250μs, communication jitter 20μs, see Section 7.2.

8.4 ROS / ROS2 Ecosystem Support

Module	Status	Description
ROS driver node	Integrable	Encapsulates ROS joint_state_publisher / trajectory_msgs via the port 7000 protocol
ROS2 driver node	Integrable	Builds ROS2 nodes based on NexDroid OpenAPI, supporting lifecycle management
URDF model	Available	Automatically generates URDF description based on DH parameters
TF tree	Integrable	Publishes joint transforms in real time → builds complete TF tree
MoveIt adaptation	Integrable	Exposes joint_trajectory_action / follow_joint_trajectory
ROS2 Control	Integrable	Interfaces with the ros2_control framework through the hardware interface layer

Note: ROS/ROS2 drivers require secondary development integration based on NexDroid OpenAPI. INEXBOT provides interface documentation and example code.

8.5 Recommended Development Toolchain

Tool	Purpose
Visual Studio / VS Code	C# host computer development
Python 3.8+	Algorithm prototyping and verification
ROS / ROS2	Robot software ecosystem integration
Lua scripts	Rapid logic writing on the teach pendant side
TCP/UDP debugging tools	Communication protocol verification

---

9. Dynamics Interface Open Specifications

9.1 Open Interfaces

Interface	Access Method	Description
Joint angle θ	Real-time read (1ms)	Current joint angle of 7 axes
Joint velocity θ̇	Real-time read (1ms)	Current joint angular velocity of 7 axes
Joint torque τ	Real-time read (1ms)	Current joint torque of 7 axes (requires servo feedback)
Joint command τ_cmd	Write	Directly issue joint torque commands in torque mode
DH parameter table	Read/Write	Standard DH / modified DH parameters, supporting runtime modification
Mass/inertia parameters	Read/Write	Mass, center of mass, inertia tensor of each link
Forward kinematics	Call	θ → TCP pose, built-in/user-defined
Inverse kinematics	Call	TCP pose → θ, built-in/user-defined
Jacobian matrix J(q)	Calculate	Geometric Jacobian / analytical Jacobian

9.2 User-Defined Algorithm Entry Points

Replacement Point	Replaceable Function	Interface Form
Joint control law	PID → sliding mode/adaptive/MPC/RL	Joint position/velocity/torque commands
Forward kinematics	Default algorithm → custom	θ → XYZ+RPY
Inverse kinematics	Default algorithm → custom	XYZ+RPY → θ
Trajectory planning	Trapezoidal/S-curve → custom	Path point sequence → time-parameterized trajectory
Dynamics feedforward	Add gravity/Coriolis compensation	τ_ff = M(q)q̈ + C(q,q̇) + G(q)

9.3 Data Acquisition and Recording

Function	Support Level
High-frequency status recording	Joint position/velocity/torque, 1ms sampling period
Host computer data stream	Port 7000 real-time push (1ms level)
Offline logs	Controller local logs, exportable for analysis, format compatible with .csv

---

10. Accuracy Calibration Solution — NexAutoCali Automatic Calibration System

The INEXBOT wheeled humanoid robot solution integrates the self-developed NexAutoCali robot automatic calibration system, providing full-parameter accuracy calibration solutions for seven-axis humanoid arms and six-axis humanoid arms, ensuring that dual arms achieve sub-millimeter repeat positioning accuracy in collaborative operations.

Detailed documentation: see NexAutoCali Automatic Calibration System.

10.1 Full-Parameter Calibration

Full-parameter calibration is currently the optimal solution for high-fidelity alignment between the robot's theoretical model and the physical entity. It supports reduction ratio calibration, zero point calibration, link length calibration, and full-parameter calibration, uncovering the equipment's own precision limits. After full-parameter calibration, TCP accuracy can be improved from 3mm to 0.5mm, with an accuracy improvement of up to 500%.

10.2 Full Coverage of Seven-Axis / Six-Axis

Adapts to seven-axis collaborative arms (SRS structure / cross-cross structure) and six-axis collaborative/industrial arms, compatible with the dual-arm independent configuration of wheeled humanoid robots, and can provide customized adaptation models according to user needs.

10.3 Full-Process Automation

Leveraging INEXBOT's drive-control advantages, the entire process from automatic measurement point planning → automatic operation → automatic laser tracker measurement → algorithm calculation → calibration result output is automated, greatly improving convenience and reliability.

10.4 Efficient and Fast

• Simplified operation logic, reducing training and usage difficulty
• Quick-mount structure design, software/hardware environment ready in 5 minutes
• Optimized calibration process, standard 50-point full-parameter calibration takes < 5 minutes

10.5 Highly Integrated All-in-One Machine

The system integrates laser tracker measurement, robot motion control, automatic calibration point/measurement point planning, fully automatic control of calibration/testing processes, and algorithm calculation into one unit. A single software can complete all calibration and measurement work without additional programming.

10.6 Full-Function Testing

• Built-in robot calibration module: reduction ratio calibration, zero point calibration, link length calibration, full-parameter calibration
• Built-in 14-item robot performance index testing module
• Fully compliant with national standard GB/T 12642—2013 / ISO 9283:1998

10.7 Multi-Brand Tracker Adaptation

Already adapted to multiple high-precision laser trackers such as API Radian Core, with more brands being continuously adapted.

10.8 Calibration Process Overview

Connect robot → Read parameters → Automatically plan measurement points → Automatically run tests → Data acquisition → Algorithm calculation → Generate calibration results

---

11. Product Packaging List

No.	Component	Quantity	Remarks
1	Controller host	1 unit	ABOX-6116, including Debian 11 / Ubuntu 20.04
2	T30 teach pendant (optional)	1 unit	Including 5m connection cable
3	Power cable	1 piece	DC 12~24V terminal interface
4	EtherCAT network cable	2 pieces	CAT5e or above
5	Technical documentation USB drive	1 piece	Including development manual SDK protocol documentation
6	NexDroid OpenAPI development kit	1 set	C# / Python example code

---

12. Related Document Public Links

The following are public access links for related documents in the INEXBOT Technology knowledge base (doc.inexbot.com):

12.1 Product Documentation

Document	Link
INEXBOT industrial robot controller specification	https://doc.inexbot.com/产品资料/控制系统/工业机器人控制器C1201
INEXBOT industrial robot controller (compact) specification	https://doc.inexbot.com/产品资料/控制系统/工业机器人控制器C1102
INEXBOT industrial robot controller (multi-axis) series specification	https://doc.inexbot.com/产品资料/控制系统/工业机器人控制器C2200系列
T30 teach pendant specification	https://doc.inexbot.com/产品资料/控制系统/T30示教器
NexAutoCali automatic calibration system	https://doc.inexbot.com/产品资料/精度标定/自动标定系统NexAutoCali
Supported servo models	https://doc.inexbot.com/技术资料/支持的伺服型号
Supported robot types	https://doc.inexbot.com/技术资料/支持的机器人类型
Supported external axis types	https://doc.inexbot.com/技术资料/支持的外部轴类型

12.2 Operation Manuals

Document	Link
Motion control instructions	https://doc.inexbot.com/操作手册/24.03版本/运动控制类指令
Robot DH parameter description	https://doc.inexbot.com/操作手册/24.03版本/机器人DH参数说明
Multi-robot and dual-robot collaboration	https://doc.inexbot.com/操作手册/24.03版本/多机与双机协作
New dual-robot functions	https://doc.inexbot.com/操作手册/24.03版本/新双机功能
Multi-robot coordination instructions	https://doc.inexbot.com/操作手册/24.03版本/多机协调类指令
Interference zones	https://doc.inexbot.com/操作手册/24.03版本/干涉区
External axis user manual	https://doc.inexbot.com/操作手册/24.03版本/外部轴使用手册
Independent axis control	https://doc.inexbot.com/操作手册/24.03版本/独立轴控制
Servo response time	https://doc.inexbot.com/操作手册/24.03版本/伺服响应时间
System function debugging manual	https://doc.inexbot.com/操作手册/24.03版本/系统功能调试手册
Calculate target pose instructions	https://doc.inexbot.com/操作手册/24.03版本/计算目标形态指令
Tool hand calibration manual	https://doc.inexbot.com/操作手册/24.03版本/工具手标定手册
User coordinate calibration manual	https://doc.inexbot.com/操作手册/24.03版本/用户坐标标定手册

12.3 Secondary Development and Communication Protocols

Document	Link
Port 7000 user manual	https://doc.inexbot.com/操作手册/24.03版本/7000端口
TCP communication function manual	https://doc.inexbot.com/操作手册/24.03版本/TCP通讯功能手册
Modbus function user manual	https://doc.inexbot.com/操作手册/24.03版本/Modbus功能使用手册
OPC-UA parameters	https://doc.inexbot.com/操作手册/24.03版本/OPC-UA参数
EIP function operation instructions	https://doc.inexbot.com/操作手册/24.03版本/EIP功能操作说明
FINSTCP user manual	https://doc.inexbot.com/操作手册/24.03版本/FINSTCP使用手册
Lua tutorial	https://doc.inexbot.com/操作手册/24.03版本/Lua教程
PC simulation software tutorial	https://doc.inexbot.com/操作手册/24.03版本/PC支持仿真软件使用教程