A self-guiding vehicle, like an autonomous car or a truck, requires a variety of different technologies to successfully navigate crowded city streets and highways.
One of the more popular is lidar (light detection and ranging), which relies on pulsing laser light beams to create a precise 3D image of the local surroundings. Radar is utilized for ranging targets using a different part of the EM spectrum, while video cameras read signs and detect color, and high-definition maps help with localization, and more. In addition, many autonomous cars also rely on GPS information transmitted from satellites and ground antennas to help provide additional navigation data.
All of these technologies involve the external environment in order to provide data back to the software stack for localization, perception, and control.
In comparison, an inertial measurement unit (IMU) sensor is different from all of the other sensors typically found in an autonomous vehicle because an IMU requires no connection or knowledge of the external world. This unique, independent property of the IMU makes it a core technology for both safety and sensor fusion.
An IMU is a device that directly measures the three linear acceleration components and the three rotational rate components (6-DOF) of a vehicle.

Figure 1: Each common navigation solution for guiding autonomous cars has problems. None can work as the only solution, yet an accurate IMU can minimize the issues in red.
Seven Reasons Your Autonomous Car Design Needs an IMU
1. A Good Attitude
An accurate IMU can precisely determine and track attitude. Yes, knowing the car’s position or location is extremely important, but when in motion, the direction or heading is equally crucial.
Dynamic control of the vehicle requires sensors with dynamic response, and an IMU does a very good job of accurately tracking dynamic attitude changes. Moreover, attitude is needed to control the vehicle and is often an input into other algorithms. While lidar and cameras can be useful in determining attitude, GPS is often pretty useless. Moreover a stable independent attitude reference has value in calibration and alignment.
2. Safety—What Is the Redundancy Plan?
The system engineer needs to consider every scenario, and always have a back-up plan.
Failure Mode Effects Analysis (FMEA) formalizes this requirement into design requirements for risk mitigation. FMEA will ask what happens if the lidar, radar, and cameras all fail at the same time? An IMU can dead-reckon for a short period of time, meaning it can independently determine full position and attitude for a short while. An IMU alone can slow the vehicle down in a controlled way and bring it to a stop, providing the best practical outcome for a bad situation. While this may seem like a contrived requirement, it turns out to be a fundamental one to a mature safety approach.
3. How to Keep the Autonomous Car Within Its Lanes
Surprisingly, when not distracted or drunk, humans are pretty good at driving. A typical driver can hold their position in a lane to better than 10 cm. This is actually very tight.
If an autonomous vehicle wanders in its lane, then it will appear to be a bad driver. As an example, during a turn, poor lane keeping could easily result in an accident.
The IMU is a key dynamic sensor to steer the vehicle dynamically, moreover the IMU can maintain a better than 30-cm accuracy level for short periods (up to 10 seconds) when other sensors go offline. The IMU is also used in algorithms that can cross compare multiple ways to determine position/location and then assign a certainty to the overall localization estimate. Without the IMU, it may be impossible to even know when the location error from a lidar solution has degraded.

Figure 2: During turns, an accurate IMU plays a key role in lane keeping.
4. Lidar Is Still Expensive
Tesla has been recognized for its “No Lidar Required” approach to autopilot technology.
If you don’t have lidar, a good IMU is even more critical because camera-based localization of the vehicle will have more frequent periods of low accuracy simply depending on what is in the camera scene or the external lighting conditions. Camera-based localization uses “SIFT” feature tracking in the captured images to compute attitude. If the camera is not stereo (often the case) inertial data from the IMU itself becomes an essential core part of the math needed to compute the car’s position and attitude.
5. High Performance Compute Power Is Not Free
The powerful combination of high-accuracy lidar and high-definition maps is at the core of the most advanced Level 4 self-driving approaches such as those being tested by Cruise and Waymo. In these systems, lidar scans are matched in real time to the HD map using convolutional signal processing techniques. Based on the match, the precise location of vehicle and attitude is estimated. This process is computationally complicated and expensive.
While we all like to believe the cost of compute is vanishingly small, on a vehicle it simply is not that cheap. The more accurately the algorithm knows its initial position and attitude, the less computation required to compute the best match. In addition, by using IMU data, the risk of the algorithm getting stuck in a local minimum of HD map data is reduced.
6. GPS/INS: Enabling Highly Accurate GPS Guidance
In most of today’s production vehicles, GPS systems utilize low-cost, single-frequency receivers, which makes the GPS accuracy relatively useless for vehicle automation.
However, low-cost, multi-frequency, network-corrected GPS is on the way from a wide variety of silicon suppliers. On top of this upcoming silicon, network correction-based solutions such as RTK and PPP can provide GPS fixes to centimeter level accuracy under ideal conditions.
However, these solutions are very sensitive to the environment—such as bridges, trees, and buildings. It is well established that the way to overcome this challenge and improve high-accuracy GPS reliability is to use high-accuracy IMU aiding at a low level in the position solution. Such GPS/INS techniques include tightly coupled and ultra-tightly coupled GPS/INS. These are coming soon to the automotive market.
7. Most Cars Already Have an IMU
Most current production automobiles already have anywhere from 1/3 of an IMU to a full IMU on board.
Vehicle stability systems rely heavily on a z-axis gyro and lateral x-y accelerometers. Roll over detection relies on a gyro mounted with its sensitive axis in the direction of travel. These sensors have been part of the vehicles safety systems for over a decade now.
The only problem is that the sensor accuracy is typically too low to be of use for the prior six use cases. So why not upgrade the vehicle to a high-accuracy IMU and let it drive independently? The main barrier is cost.
However, open source resources make it easy for engineers to quickly develop these solutions. You don’t need a PhD or decades of IMU development experience in order to develop localization or navigation systems for autonomous cars and trucks.