Adaptive and Robust Techniques for Robot Perception
Being able to understand the environment (usually time-varying and unknown a priori) is an essential prerequisite for intelligent/autonomous systems such as intelligent mobile robots. The environmental information can be acquired through various sensors, but the raw information from sensors are often noisy, imprecise, incomplete, and even superficial. To obtain from raw sensor data an accurate internal representation of the environment, or a digital map with accurate positions, headings, identities of the objects in the environment, is very critical but very difficult in the development of robotic systems. The major challenge is from the uncertainty of the environment and the insufficiency of sensors. Basically there are two categories of techniques for handling uncertainties: adaptive and robust. Adaptive techniques exploit a posteriori uncertainty information that is “learnt” on-line, whilst robust techniques take advantage of a priori knowledge about the environment and sensors. We are mainly interested in model-based approaches. We are investigating techniques for automatic error detection and error-driven model adaptation or parameter adjustment. We are also developing multisensor data fusion methods and multiple model approaches, including complex task decomposition, individual model design, and intelligent model switching or fusion.
This project was partly supported by the Research Promotion Fund of the University of Essex.