Learning Long-Range Perception Using Self-Supervision from Short-Range Sensors and Odometry
This work addresses obstacle detection for mobile robots, but it is incremental as it builds on existing self-supervised and neural network methods.
The paper tackles the problem of predicting short-range sensor outputs from long-range sensor data using a self-supervised approach, implemented on a mobile robot to detect obstacles with a convolutional neural network, achieving quantitative evaluation on unseen scenarios and demonstration in obstacle-avoidance control.
We introduce a general self-supervised approach to predict the future outputs of a short-range sensor (such as a proximity sensor) given the current outputs of a long-range sensor (such as a camera); we assume that the former is directly related to some piece of information to be perceived (such as the presence of an obstacle in a given position), whereas the latter is information-rich but hard to interpret directly. We instantiate and implement the approach on a small mobile robot to detect obstacles at various distances using the video stream of the robot's forward-pointing camera, by training a convolutional neural network on automatically-acquired datasets. We quantitatively evaluate the quality of the predictions on unseen scenarios, qualitatively evaluate robustness to different operating conditions, and demonstrate usage as the sole input of an obstacle-avoidance controller. We additionally instantiate the approach on a different simulated scenario with complementary characteristics, to exemplify the generality of our contribution.