Robots are essentially machines capable of autonomous operation. To achieve autonomy, a comprehensive suite of sensors is needed to meet the requirements of various tasks, such as autonomous navigation, object detection, and proximity sensing. As sensor technology has matured, the cost of various sensors has gradually decreased over the past few years, driving their wider application in the robotics field. As highly integrated machines, robots require the use of numerous sensors, including optical encoders, current sensors, inertial sensors, cameras, and LiDAR (Light Detection and Ranging).
Based on the data they sense, robot sensors can be broadly categorized into two types: proprioceptive sensors and external sensing sensors. Proprioceptive sensors collect internal data such as the robot's speed, torque, and position. These sensors are typically used for robot control. Conversely, external sensing sensors primarily collect data about the external environment, sensing environmental parameters such as the location of obstacles, external forces applied to the robot, and many other inputs.
Robot sensor classification
Tactile sensors, cameras, LiDAR, radar, and ultrasonic sensors are some typical examples of external sensing sensors. With the increasing prevalence of various robots and the growing market demand for "intelligent" robots, IDTechEx, a well-known UK research company, predicts that the robot sensor market will experience rapid growth over the next 20 years.
According to MEMS Consulting, IDTechEx's latest report delves into nine common sensor types, nine types of robots, and 29 applications, providing in-depth analysis of key enabling technologies, manufacturers, and markets, and offering accurate forecasts of market size and sales trends over the next 20 years.
This report studies the types of robots, sensors, and their tasks.
Navigation and mapping sensors
Thanks to the development of autonomous driving technology, autonomous robot operation has made tremendous progress in the past decade. As one of the most important manifestations of robot autonomy, autonomous operation enables robots to move independently with minimal human intervention and complete many tasks such as logistics delivery, agricultural applications, surveying, and exploration.
Robot 3D sensor applications
Autonomous operation involves two steps: mapping and navigation. Initially, the robot needs to be able to map its surroundings, build an environmental model composed of point clouds, and plan its movement trajectory/path. Then, it uses navigation sensors to locate and operate along the planned trajectory. Both steps require sensors for object detection, navigation, and collecting data from the surrounding environment. In practice, depending on the different working environments, various navigation and mapping sensors often need to be used together, and sensor fusion algorithms are used to process the data from different sensors.
Typical navigation and mapping sensors include LiDAR, radar, cameras, GPS/GNSS, and ultrasonic sensors (including 1D and 3D). The table below compares the advantages and disadvantages of some of these sensors, and IDTechEx provides a more in-depth technical analysis in its report.
Comparative Analysis of Typical Navigation and Mapping Sensors
Collision/proximity sensor
Beyond autonomous operation, safety remains the top priority for all robots, especially with the increasing sophistication of human-robot interaction (HRI) and the growing complexity of tasks. IDTechEx anticipates that regulations will become increasingly stringent to ensure safe levels of HRI. For robots to meet safety requirements, they need to be able to accurately perceive the relative distance and potential collisions with human operators. When a human operator/object approaches, the robot needs to slow down or stop. This typically requires the use of collision/proximity sensors.
With advancements in sensor technology, the line between collision detection and proximity detection has become blurred. The fundamental difference between collision detection and proximity detection lies in the distance between the object and the sensor/robot. From a technical perspective, collision/proximity sensors are typically based on one or more of five detection principles: optical reflection, time-of-flight, triangulation, capacitance measurement, and ultrasonic ranging.
The diagram below compares the detection principles of several commercial sensors, outlining the response time and maximum detection range of each method. Overall, robot end-users desire sensors with fast response times, large detection ranges, and small footprints. However, certain characteristics may be traded off for different applications. For example, indoor autonomous robots used for logistics or material handling may not require the large detection range of outdoor mobile robots.
Comparative Analysis of Proximity Detection Technologies
Summarize
IDTechEx's in-depth analysis of sensors in 29 robotic applications indicates that the robotic sensor market is poised for rapid growth. Given the enormous market size of robots and autonomous machinery, IDTechEx projects the robotic sensor market will grow 20-fold over the next 20 years, reaching over $80 billion by 2043. Each type of robot and its sensors has its own specific needs and market drivers. This massive market size and rapid growth represent numerous opportunities, which this report analyzes in detail.
Robot sensor market forecast by application (sample document blurred)
