Thanks to the maturity of new sensors, the penetration rates of L2 and L3 in new smart cars sold in China are 35% and 9% respectively in 2022, 51% and 20% respectively in 2023. Every L3 smart car needs to be equipped with at least one high precision inertial measurement unit.
Importance of Inertial Measurement Unit
Why IMU is so important for autonomous vehicle?
The most important points required for the success of autonomous vehicles are the real-time, continuous, and accurate position and movement of the vehicle during any environmental change. All types of autonomous vehicles such as cars, trucks, buses, drones, robotic systems must be able to determine their position and orientation in all weather, temperature, environments, and without GNSS signals or signal attenuation.
What role does IMU play in autonomous vehicle?
In high-level autonomous driving, inertial measurement unit will act as a safety line to ensure the safety of autonomous driving. For example, in tunnels or bad weather, navigation systems fail, or sensors such as LIDAR or cameras fail, inertial measurement units can estimate the path to ensure the normal operation of autonomous driving. While more and more cars use high-level autonomous driving such as L3 and L4 , inertial measurement unit will play a more important role. It uses dead reckoning to ensure the safety of automatic driving when the satellite signal is blocked or lost
Misunderstanding for IMU’s precision and resolution
Inertial measurement unit is a new type of sensor. There are a lot of good and bad articles about inertial measurement unit on the Internet, so that some customers misunderstood the concept of precision and resolution.
IMU Precision
Precision is the maximum difference value between the measured value of the sensor and the true value, also known as the measurement error. The smaller the value, the smaller the error. The inertial measurement unit integrates an accelerometer and MEMS gyroscope sensors. It can also integrate a magnetometer. Common ones include 3-axis, 6-axis, and 9-axis. Integrated magnetometers are usually not needed in the field of autonomous driving.
Precision metrics of accelerometers and gyroscopes
The certainty error can be corrected by calibration and compensation, and the uncertainty part determines the precision. The gyroscope and accelerometer parameters of an inertial measurement unit are usually listed separately, and the more commonly used metrics are zero bias instability of IMU sensors, random walk, and noise density.
For gyroscopes, zero bias instability has long been regarded as the absolute standard for gyroscope parameter. It is the lower limit for describing the performance of gyroscopes. However, zero bias (in) stability is only meaningful for medium and high performance inertial sensors. Low precision gyroscope with a little rotation will produce a huge error. zero bias (in) stability is not much significant.
The zero biased instability metrics follows a normal distribution and typically uses typical values (mean values) and maximum values (1σ).
IMU Resolution
Resolution is the minimum measurement that leads to a change in the indicated value, or the ability of the sensor to feel the smallest change in the measured value. The minimum measurement is not related to precision.
For example, an inertial chip is designed with a resolution of 256 LSB/°. This means that the inertial chip is designed to divide each degree into 256 smallest measurements (also known as the Least Significant Bit).
The reciprocal of the resolution is the sensitivity. Take the inertial chip as an example, the sensitivity =1/256=0.0039°/LSB. Sensitivity is artificially defined at design time.
Precision determines IMU performance.
The inertial measurement is generally installed on the center of gravity of the measured object. The gyroscope and accelerometer are the main components. They measure the angular velocity and acceleration respectively. The precision actually refers to the parameters measured by gyroscopes and accelerometers. Their numerical value directly reflects the precision of the inertial navigation system.
The parameters of the gyroscope include zero bias instability (Z axis), zero bias instability (X/Y), zero bias at full temperature, Angular random walk, etc. The parameters of the accelerometer include: zero bias instability, zero bias at full temperature, measurement range, and velocity random walk.
Then, from the point of view of the user to select and use the sensor, it is not necessary to consider the resolution, just look at the precision.
DaischSensor’s High Precision IMU
As automotive grade processor chips and MEMS inertial components, DaischSensor’s IMU has the following features, including full temperature calibration, fast integration interface for mass production process requirements, excellent reliability performance, very low zero bias noise and high performance motion perception in the full temperature range (-40° ~ 85°).
As the tactical inertial measurement unit of Daisch, IM1A is designed for high performance motion perception under various conditions. At the same time, according to the precision requirements of different customers, Daisch has made different matching models, covering nearly 5 models of high, medium and low precision, such as IM1A2690 IM1A2685 / IM1A2680 / IM1A2670 / IM1A2660.
IM1A can be used in the automotive field, construction machinery, AGV, UAV, high-precision navigation, etc.
The IM1A has six degrees of freedom motion perception. They integrate a three-axis MEMS accelerometer and a three-axis MEMS gyroscope in a unique redundant design that maximizes performance. In terms of performance, the IM1A have excellent gyroscope and accelerometer zero bias instability of 1.5°/h and 25µg. This enables long-term localization dead reckoning and maintains excellent navigation performance.
