Kalman filters are used in robotics to correct measurement errors. Imagine trying to precisely predict the position of an outdoor robot. The robot is equipped with a GPS and is able to measure the speed of its wheels (odometry). Only using GPS leads to measurements that are not precise while using only odometry leads to increasingly wrong estimates. Instead, what a Kalman filter does is fuse the information from odometry with the GPS measurements. This is done by, at each step of the robot control, predicting future sensor readings based on the commands given to the robot. The difference between the predicted sensor readings and the actual sensor readings is then used to update the filter. In this manner, the robot is able to improve its position estimate over time.

In work by Huang et al., groups of indoor robots attempt to estimate their global position and orientation using a special type of Kalman filter called the “Extended Kalman Filter”. Since they do not have access to GPS or landmarks in the environment, robots “cooperatively localize” by using odometry and measuring their relative position to neighboring robots. However, Kalman filters can be challenged when the measurements they make do not give them enough information with respect to what they are trying to predict. For example, sensor measurements might only provide meaningful information to correct position estimates but not global orientation. In these cases, the system is “not observable” and the Kalman filter can result in inconsistencies.

To overcome this challenge Huang et al. propose two ways of extending Kalman filters so as to constrain the observability of the system. Results are given in simulation and using four Pioneer I robots that were able to successfully estimate their pose. Odometry measurements were derived from wheel encoders and relative position was computed using an overhead camera thanks to the rectangular tags on each robot shown in the figure below. Results show that both developed algorithms outperform standard extended Kalman filters.

In the future, researchers hope to extend their approach to 3D localization.

Related posts

• SLAM with aerial images
• RFID-based global positioning

Having a robot figure out its global position is required in many real world applications, it’s also one of the biggest challenges in robotics.

The easiest approach is to have a robot blindly keep track of its movements (odometry) from a known starting position. Odometry alone however quickly results in an add-up of errors that make the localization unusable.

To help the robot along the way, Boccadoro et al. propose to place passive Radio-Frequency IDentification (RFID) tags in known positions in the environment. These smart tags are interesting because they are typically low cost and require no energy to function. Robots equipped with RFID readers can detect a tag within a 1m range, although with a lot of noise. Algorithms are then needed to combine the robot’s sensors, in this case odometry, with the noisy RFID readings to precisely estimate its global position.

For this purpose, two types of Kalman Filters are implemented and compared to a Particle Filter method that typically has much larger computational cost. Experiments were conducted using a Pioneer P3-DX driving around a corridor equipped with 6 RFID tags.

Results show that the first method is fast but imprecise when tags are sparse (figure left). The second approach has higher computation requirements than the first but is able to obtain estimates as good as the Particle Filter method with few tags (figure right).

The path reconstructed through the various methods proposed: a red line is used to represent the estimation of the second loop of the robot path, the green line is used for the last loop; the line in blue is ground truth.

In the future, authors hope to investigate the optimal placement of RFID tags to achieve even better position estimates.

Related posts

• Cooperative localization using Kalman filtering
• SLAM with aerial images