RSJ データ工学ロボティクス研究専門委員会、NEDO特別講座共催 8/5(金)公開講演会の案内

講師:Thomas Westfechtel(東京大学)

Generating high-level knowledge enriched maps through observing the environment and its usage over time.

One of the bottlenecks for autonomous robots is the overly complex structure of human-made environments. The robot has to perceive and understand its environment to a high degree. In order for the robot to react to and interact with the environment, high-level knowledge of the environment is required. To alleviate this problem semantically enhanced maps are often employed (i.e. for self-driving vehicles these maps include driving lanes, traffic signs, and lights, …). While such kind of high-level knowledge-enriched maps proves to be very effective, the generation of these maps is tedious. In the lecture, we exploit that the environments are in active use. By observing their usage over time we extract high-level knowledge and enrich environmental maps with it in order to generate a better understanding of the environment for the robot.

スケジュール:8/5(金)16 :00-17:00 公開講演






Leave a Response