Nowadays robotics are developing rapidly, from automated devices that perform repetitive actions in specific positions, transforming to intelligent devices that capable of sensing, processing, decision-making, and executing in complex environments and jointed actions.
This trend is driving a fundamental transformation in robotics, leading to diverse forms and joint functions, across various and specific applications.
I. Development Directions: Three Core Functional Directions
If robotics in old times were automated devices that performing with programs, then the robotics in future will focus on more complex and flexible logics. This is specifically reflected in three directions:
– From repetitive actions to more flexible & diverse functions: information processing and execution.
Old time robotics relied on pre-setted programs to perform, in certain positions, nowadays robotics are turning to operating systems, that driven by algorithms, giving them spatial memory (remembering the position of objects, even after left the vision field) and the ability to reason, based on common sense (understanding causal relationships).
For example, at the World Robot Conference, the humanoid robot Galbot G1 was able to adjust it’s plans immediately, and continue the entire task of making breakfast when breads and kettle were temporarily moved.
– More refined and sensitive structures, from rough movements to tactile perception: The operational precision of robotics is evolving from “recognition” to “precise detail.”
Through sensing and millisecond-level of touch-force control, robotics could not only perform precision assembly and record tactile pressure accurately in factories, but also complete extremely precise actions, such as sewing a grape(medical level, for surgery) and shaving human beards.
– From closed to open environments, from specialized to general-purpose: This is the core trend and requires a lot more intelligent functions.
Robotics are moving out from factories and moving into flexible environments, such as homes, shopping malls, entertainments, and disaster areas, integrating algorithms and intelligent functions, compatible for variable missions in diverse scenarios.
For example, a robot was showcased by Unitree Robotics, it can switch from bipedal to quadrupedal modes, changing it’s locomotion to move in different terrains for various missions.
II. Current and Future Forms of Robotics
Based on the directions above, we may see the following types of robots now and in the near future:
| Category | Main Form | Typical Application Scenarios | Core Capabilities / Characteristics |
|---|---|---|---|
| Industrial Robotics | Robotic arms, humanoid robots, hybrid robots | – Mechanical manufacturing- Precision assembly – Welding and painting- Operations in hazardous chemical environments – Inspection and maintenance in high-risk environments | – High precision, efficiency, and repeatability – Designed to replace humans in dangerous or chemically hazardous workplaces – Capable of operating in complex industrial environments – Example: Wheeled hybrid robots that can navigate complex terrain, carry up to 350 kg, and operate in steel refineries, chemical workshops, and nuclear industry environments |
| Service / Consumer Robotics | Humanoid robots, wheeled robots, quadruped robots | – Home services (floor cleaning, laundry, cooking) – Retail and catering services (coffee making, bartending, dessert preparation, delivery) – Hotel delivery services – Elderly care – Pet care | – Focus on human–robot interaction and safety – Stronger generalization capabilities for diverse tasks – Ability to perform complex, long-duration tasks in human environments – Designed for close collaboration with people |
| Specialty Robots | Quadruped robotic dogs, tracked/wheeled platform robots, biomimetic robots | – Disaster emergency rescue – Military reconnaissance – Deep-sea exploration – Space exploration – Operations in extreme environments | – High environmental adaptability and mobility – Designed for challenging and unpredictable conditions – Capable of autonomous operation in complex environments – Example: Quadruped robotic dogs equipped with robotic arms that can perform tasks such as pressing elevator buttons, waste disposal, and flexible delivery operations |
General-purpose robotics, specifically humanoid robots, can theoretically compatible with any human work & life scenario. The ultimate goal is to possess general intelligence and adaptabilities like human beings. The founder of Unitree Robotics, predicts that in two to three years, or even five to ten years, when robotics could complete 80% of household tasks, the industry will experience it’s “ChatGPT moment.”
In terms of industry scale, China has become the world’s largest robotics market. In the first 6 months of 2026, the delivery amount of humanoid robotics had exceeded 40,000 units, and the sales revenue of global robotics hardware is predicted to grow at an average annual rate as 38%, from 2025 to 2030.
III. Application of PCBs on Robotics
The internal structure of robotics is complex with various functions, that requires various types of PCB(printed circuit board) to perform various tasks and assignments, achieving systematically integration and collaboration.
These PCBs like the robotics “brain” (operating system and signal processing), “nerves” (signal transmission), and “muscles” (drive output), collectively constituting every robotic systems.
Robotics PCBs are mainly divided into four modules: main control board, driver board, sensor board, and power management board.
Various types of PCBs take different functions, tasks, require different production technologies and facing various challenges.
1. Main Control Board: The Robotics “Brain”
– Processes data from various sensors, responsible for information processing, action planning, multi-sensor combination, and output signals. It is the core of computation and decision-making.
– Typically multilayer HDI board (8-24 layers or higher), employing arbitrary layer interconnection and micro-blind via technology, carrying high-performance chips in large-size, such as MCU in BGA packages.
– Ultra-high density and signal integrity:
trace width/spacing can be as fine as 2mil (0.05mm), and inter-layer alignment accuracy must be controlled within ±15μm.
Strict impedance control is required for differential signals (such as PCIe, DDR), typically within 100Ω ±5% to ensure lossless transmission of high-speed signals.
– Efficient heat dissipation: High-performance chips generate concentrated heat, requiring rapid heat dissipation through thermal via arrays and embedded copper blocks.
Core Components:
– Processor: Typically a multi-core heterogeneous SoC, such as the NVIDIA Jetson series or Qualcomm RB5 platform, integrating a CPU, GPU, DSP, and a dedicated AI accelerator (NPU).
– High-speed Memory and Storage: Equipped with high-bandwidth memory such as LPDDR5 or GDDR6, and UFS or NVMe SSDs for fast boot.
– Key Interfaces: Requires high-speed data interfaces, such as PCIe (for connecting AI accelerator cards), MIPI (for connecting cameras), and Gigabit/10 Gigabit Ethernet (for connecting LiDAR or host computers).
Challenges during Design and Manufacturing:
– High-Density Interconnect PCB (HDI PCB): The main control board is typically an 8-24 layer or higher HDI board, employing micro-blind via technology.
To ensure signal quality, the impedance control of critical differential signal lines (such as PCIe and USB) needs to be extremely precise (tolerance ±5%).
– Heat dissipation is a core challenge: GPUs and AI chips generate enormous amounts of heat.
In terms of design, in addition to adding heat sinks and fans, the PCB itself will use a metal substrate or an array of thermal vias, directly beneath the chip, utilizing the high thermal conductivity of copper to conduct heat away quickly.
– Power Integrity: The core processor experiences large instantaneous current fluctuations, therefore a complex multiphase power supply module is deployed on the board.
By using a large number of low ESR (Equivalent Series Resistance) MLCCs (Multilayer Ceramic Chip Capacitors) to filter noise and ensure voltage stability.
2. Driver Board: The Robotics “Nerves and Muscles”
– Responsible for controlling motor operation precisely and processing high-current power signals, it is crucial for the robotics to perform actions.
– Commonly uses a combination of thick copper PCB and multilayer PCBs, with independent power layers that carrying high currents, signal layers to transmitting control signals, and ground layers for isolation and reduce interference.
– High Current and High Reliability: It needs to withstand the high current required for motor drive (e.g., copper thickness of power layer could reach 3OZ, to carry currents of 15 A/mm² or even more).
In harsh environments, resin plugging process is required to fill vias and prevent shorts, corrosion, that caused by vibration, oily contamination, and temperature changes, ensuring stable operation.
Core Components:
– Driver Chip (MOSFET): Applies power devices, such as GaN (Gallium Nitride) or SiC (Silicon Carbide) for high-efficiency current control, through high-frequency switching.
– Control Unit (MCU/DSP): Typically applies high-real-time MCUs such as TI’s C2000 series or STM32, responsible for executing closed-loop control algorithms, for current, speed, and position loops.
– Current and Position Detection: Includes high-precision sampling resistors, and current-sense amplifiers for phase current detection, as well as interfaces for connecting magnetic or photoelectric encoders, to sense the motor position.
Challenges during Design and Manufacturing:
– High Current Carrying and Heat Dissipation: The copper foil on the power path needs to be very thick (2OZ to 4OZ or even thicker to 6OZ).
High-power devices often with large thermal pads on the back, conducting heat to the back heatsink through PCB vias.
– High and Low Voltage Isolation: Driver boards typically contain high-voltage (motor side) and low-voltage (control logic side).
For safety, the safety clearance must meet regulations, and the PCB itself often uses materials with a higher CTI (Current Tracking Index) rating, to prevent high-voltage creepage.
– Electromagnetic Compatibility (EMC): High-frequency, high-current switches are strong sources of interference.
Uses multi-layer PCBs to provide a complete ground plane, to reduce loop area, and adds filtering circuits at the ends of power input and motor output, to pass rigorous EMC testing.
3. Sensor/Interface Board: The Robotics “Senses”
– Responsible for connecting and processing various sensor signals, connecting cameras, LiDAR, microphones, etc., and processing massive amounts of sensor data.
– Often uses high-density interconnect (HDI PCB), sometimes also using rigid-flex PCB boards for 3D wiring in confined spaces, or for flexible connections in areas that requiring bending, such as robot joints.
– Miniaturization and High Density: To fit within the compact space of the robotics head, power management, multiple sensors, and filtering circuits, the PCBs often have to be integrated into a very small space.
– Special Materials and Processes: High-frequency PCBs may be required, to support high-frequency signal transmission (e.g., Wi-Fi, radar).
Core Components:
– Sensor Interfaces: Includes various interfaces such as cameras (MIPI-CSI), lidar (Ethernet/SPI), microphones (I2S), and IMUs (I2C/SPI).
– Signal Conditioning and Conversion: Includes an analog front-end (AFE) for signal amplification and filtering, and a high-precision ADC (analog-to-digital converter) to convert analog signals (e.g., force, temperature, sound) into digital signals.
– Coprocessor (Optional): Sometimes a low-power MCU is included to perform simple preprocessing of raw data, such as filtering or feature extraction, reducing the burden on the main controller.
Challenges during Design and Manufacturing:
– Miniaturization and Integration: To fit in compact spaces such as robotic heads, the sensor boards need to be very small (e.g., the size of a coin).
High-density interconnects (HDI PCB) and rigid-flex PCBs are often used to achieve three-dimensional connectivity.
– Signal Integrity and Noise Suppression: High-frequency camera signals and weak analog signals (e.g., microphone signals) are extremely sensitive to noise.
That requires physical isolation between sensitive analog and digital circuits, and may employ differential signal routing to resist common-mode interference.
– PCB traces for specific high-frequency sensors (such as millimeter-wave radar), require high-frequency substrates (such as the Rogers series RO4003C), and precise impedance control for microstrip or stripline lines.
4. Power Management Board
Responsible for converting, distributing, and managing power from batteries or external power supplies, providing stable power to various components.
It is a multi-layer PCB or thick copper PCB, ensuring low-impedance, high-efficiency current transmission while balancing high current and thermal management.
Besides the various functions mentioned above, robotics PCBs also face some common and severe challenges, which is the fundamental difference between them and ordinary consumer electronics PCBs:
1) Diverse types and complex processes:
A single robotics system may need to integrate three, four, or even more than ten different types of PCBs for various purposes: control, drive, sensing, and power supply, ranging from high-end main control boards to flexible sensor boards.
Each board may apply different materials, layer counts, and process requirements, bring up significant challenges to the R&D team and supply chain management.
2) “One-stop” capability is crucial:
R&D progress for robotics are long and iterative. From prototype verification to mass production, PCB suppliers need to be able to respond quickly with prototyping, while ensuring consistent quality during mass production.
Therefore, “one-stop” PCB&PCBA supplier that can cover prototyping needs for “multiple varieties, pilot run to small batches, and quick-turn, short lead times” while also undertaking mass production is critical to R&D efficiency.
3) Reliability and environmental adaptability:
Robotics often operate under high-intensity conditions, such as vibration, temperature changes, humidity, and even oil contamination.
Therefore, the PCBs must undergo a series of rigorous environmental stress tests during manufacturing, including extreme temperature cycling, humidity, vibration, and shock, to ensure long-term operational stability.
Besides these hard requirements, the design and manufacturing of robotics PCBs also have to balance multiple aspects, such as high-density interconnect (HDI PCB), signal integrity (SI), power integrity (PI), and manufacturability (DFM), making it a systematic technical engineering project.
Summary and Trends
– The technical challenges of the main control board are the integrity of high-speed signals and heat dissipation for high power consumption.
– The technical challenges of driver boards are power density, thermal management, and electromagnetic compatibility.
– The technical challenges of sensor boards are miniaturization, high integration, and maintaining weak signal fidelity under various interferences.
The trend is that, to meet the size and bandwidth requirements of high-performance robotics, these four types of PCBs are moving towards high integration. For example, integrating driver circuits and control circuits onto a single PCB, or directly integrating sensor data processing units into sensor modules.
Nowaday, Kyrid is keeping pace with the mainstream of robotics development and develop it’s ability of manufacturing and supply chain integration as well: improve the processing requirements of PCB fabrication from conventional FR-4 PCB to HDI PCB, from heavy copper PCB to rigid-flex PCB, also built great relationship with various components suppliers, not only ensure the supplying of PCB, but also help to source components.
If you are on the development progress of robotics, Kyrid surely is a reliable supplier and please reach out if you want to know more of us, such as quote and presentation.
