Mobile and ROS robots

Tankbot

An ESP32 tracked robot with a 6DOF arm on a suspension chassis; the Advanced Kit adds AI vision and voice.

Arduino
ESP32
Suggested for secondary
Suggested for university

Tankbot is an ESP32-driven robot that Hiwonder built for embedded development and large AI model projects. A 6DOF robotic arm, five joints plus the gripper, sits on a torsion spring suspension tank chassis, and Hiwonder's inverse kinematics lets it grip objects from simple coordinate inputs. Bluetooth 5.0, an RGB ultrasonic sensor and an 8-channel line follower cover app control, distance tracking, obstacle avoidance, complex line following and intersection recognition. The Advanced Kit adds the WonderLLM AI module with a 2MP camera, microphone and speaker, linked to the XiaoZhi platform for conversation, voice control and visual understanding, plus a dot matrix module and an accelerometer. It is programmed in the Arduino IDE or Keil5, with source code and schematics provided.

Zeta42 is an authorised Hiwonder partner in the Middle East.

Configurations

Choose how it ships.

Model: Standard Kit · Advanced Kit

What it does

Built for.

Arm on a suspension chassis
20 kg serial bus servos with feedback drive the arm through an open-source servo controller. The tank chassis has torsion spring suspension, anti-slip tracks and Hall encoder motors.
Sensors
The RGB ultrasonic sensor measures distance and avoids obstacles, and the 8-channel line follower handles complex lines and recognises intersections.
ESP32-S3 processor
An Xtensa 32-bit LX7 dual-core ESP32-S3 handles AI applications and multitasking, with 600 DMIPS of processing power.
Large AI models (Advanced Kit)
Through the XiaoZhi platform, Tankbot chats, follows voice commands and analyses images. A language model turns spoken requests into arm movements that grasp, move and sort objects, and at intersections the camera reads road signs for model autonomous driving.
AI vision (Advanced Kit)
The 2MP camera streams video over Wi-Fi and handles colour recognition, face recognition and target tracking. The dot matrix module and accelerometer add sensor projects such as acceleration control and dot matrix displays.
Control and development
Run it from the app, from custom code on a PC or with a PS2 wireless controller. Hiwonder supplies source code, schematics, PC software and the Android app's source code.

Where it is used

Applications.

Embedded programming courses on ESP32 in the Arduino IDE
Teaching inverse kinematics with a mobile robotic arm
Line following, intersection and obstacle challenges
Voice-controlled sorting and transport with large AI models (Advanced Kit)

Videos

See it run.

Specifications

The full sheet.

Chassis type: Suspended tank chassis
Dimensions: 306 × 246 × 560 mm
Weight: 2.8 kg (Standard Kit)
Material: Full-metal aluminum alloy chassis with anodized finish
Motor: 528 hall encoder geared motor
Encoder: AB dual-phase high-precision quadrature encoder
DOF: 5 + 1 DOF
Controller: ESP32 core module + open-source servo controller
Sensors: RGB ultrasonic sensor, 8-channel line follower sensor
Camera / audio input: WonderLLM AI module with 2MP camera, microphone and speaker (Advanced Kit)
Sensor expansion pack: Dot matrix module and accelerometer sensor (Advanced Kit)
Communication: Bluetooth 5.0 / Type-C serial port
Control methods: PS2 controller / Android app / iOS app / PC control
Programming software: Arduino IDE / Keil5
Secondary development: Source code and schematics, PC software, Android app source code
Power supply: 7.4V 2200mAh lithium battery with protection board, 60 min battery life, which varies by operational state
Tutorials: Development manual, source code, system image, and supporting software
Package specs: Approx. 3.8kg, 455 × 325 × 170 mm