Support for the AI Vision Sensor in VEX V5 and EXP Teacher Training
By Carnegie Mellon Robotics Academy
We’re excited to announce a major upgrade to our onsite and online educator training programs for VEX V5 and VEX EXP robotics platforms: we now support the VEX AI Vision Sensor! This enhancement brings real-world computer vision capabilities into the classroom and competition field, helping students and educators engage with robotics at a deeper and modernized level.
We have four incredible opportunities scheduled this year for you to learn how to bring AI into your VEX Robotics classrooms and competition teams:
- Online Training, June 23rd - 27th (2-hour daily live classes, sessions are recorded)
- Onsite Training in Pittsburgh, July 7th - 11th
- Onsite Training in Pittsburgh, July 28th - August 1st
- Online Training September 22nd - October 20th (2-hour live classes every Monday, sessions are recorded)
Why Upgrade to the VEX AI Vision Sensor?
The VEX AI Vision Sensor represents a major evolution in robot perception for the VEX Robotics system. While the now-discontinued VEX Vision Sensor focused on detecting colors, the AI Vision Sensor enables robots to go further by letting them perceive what they’re sensing.
With real-time object classification and AprilTag detection, the AI Vision Sensor empowers robots to understand their environment and make informed decisions—moving beyond simple reactive behaviors to a new level of autonomous capability.
Sense. Perceive. Plan. Act.
Modern robotics is built on a cycle of intelligent behavior:
Sense → Perceive → Plan → Act.
- Sense: Collect data (e.g., camera feed)
- Perceive: Interpret that data meaningfully (e.g., "this is an AprilTag labeled '3'")
- Plan: Decide on a goal-oriented action (e.g., navigate to AprilTag 3)
- Act: Execute behavior (e.g., align and drive toward the tag)
By embedding these capabilities into our training, we move robotics education closer to the standards used in real-world autonomous systems.
What Are AprilTags, and Why Do They Matter?
AprilTags are high-contrast visual markers that encode both an ID number and orientation—essentially QR codes designed specifically for robots. They offer reliable identification at a distance and under various lighting conditions, provide precise pose estimation including distance and angle, and support consistent localization, which is crucial for accurate navigation and interaction within a defined space.
This is a major step forward for VEX competition teams, especially those competing in VEX Robotics Competition (VRC), where autonomous routines can be the difference between winning and losing. AprilTags and AI-based detection allow for more robust, repeatable autonomous behavior—a competitive edge for advanced teams.
Supporting Legacy Vision Sensors
We recognize that many classrooms and teams still use the earlier VEX Vision Sensor, and we’re committed to supporting those systems. All existing curriculum resources, tutorials, and documentation for the Vision Sensor will remain available.