Dual-Processor Design Approaches to Distributed Processing in Embedded Robotics

Overview

Conventional embedded robotics microcontrollers lack the computational capacity for modern real-time perception and control, facing clock speed and memory bottlenecks when used for tasks like real-time sensor fusion or path planning. Such microcontrollers often underutilize the low-level microcontroller, leading to a lack of functionality. This research introduces and evaluates dual-MCU architectures designed to overcome these constraints. Two dual-MCU boards were designed and fabricated, featuring a base ESP32-S3 chip for low-level control tasks, and a separate second microcontroller (RP2354B/Kendryte K210) for robotics and machine learning tasks. Key technical innovations include the hardware flow controlled UART communication system, dynamic task migration, the onboard sensor suite, and the integrated USB hub architecture which eliminated the need for physical selection switches or dual USB-C. Performance evaluation of motor control, computer vision, and multi-sensor fusion demonstrated up to 22% improvements in real-time latency, and improved processing speeds when compared to single-MCU boards. These findings establish design principles for dual-MCU systems to close the performance gap between low-cost consumer microcontrollers and high-end robotics processors.

Competition history

• AJAS 2026 Category not listed

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Browse more like this

• All Embedded Systems projects
• AJAS Embedded Systems projects
• Every AJAS 2026 project