A Modular Self-Driving System Breaks Driving Into Easier Tasks
A self-driving pipeline can divide the problem into three main parts: detect nearby vehicles, detect pedestrians, and choose a driving path. Because each part is narrower than the full driving problem, it can often be learned with less labeled data than a single end-to-end model.
0
1
Tags
Machine Learning
Deep Learning
Supervised Learning
Dive into Deep Learning @ D2L
Data Science
Machine Learning Strategy
Machine Learning Yearning @ DeepLearning.AI
Related
A Modular Self-Driving System Breaks Driving Into Easier Tasks
Why can a self-driving system perform much worse than a human driver even if each module seems strong on its own?
True or False: In a delivery robot system, every component, including route selection, must be implemented as a learned model.
Adding a _____ detector to a bridge inspection workflow can give the maintenance team damage-location information without requiring them to inspect raw drone photos.
Which three parts make up the simple warehouse-robot navigation pipeline described here?
True or False: Every stage in a machine-learning system must be learned from data.
A warehouse navigation system underperforms overall even though localization, mapping, and obstacle detection are near human-level. Adding a _____ detector can provide the missing information.
Match each warehouse robot pipeline component to its role in the pipeline.
Arrange the steps for diagnosing a multi-stage document scanning system that lags behind people even though each module seems strong.
Why a Strong Set of Modules Can Still Produce Weak End-to-End Results
True or False: In this setting, collecting labeled data for road-sign and vehicle detectors is relatively easy because public datasets and crowdsourcing can supply many examples.
Scaling Labels With Distributed Human Annotation
Match each robotics pipeline diagnosis to its correct description.
Order the steps for fixing a missing-information problem in a warehouse robot system by adding a pallet detector.
Explain why a navigation stack can fail even when its parts look strong.
Find the bottleneck in a warehouse navigation pipeline.
Explain why adding a detector can simplify a processing pipeline.
Learn After
Main Modules in a Modular Self-Driving System
True or False: In a self-driving system broken into stages, each stage is usually a simpler learning problem and can often be trained with less data than a single end-to-end model.
A warehouse robot workflow is divided into four main stages: scan the shelf, choose a path, travel to the target, and confirm the pickup. How many main stages are there?
Which three subtasks are the core stages of a self-driving navigation pipeline?
A modular speech-recognition pipeline usually needs more labeled data for each subtask than a single end-to-end model would need.
A delivery-drone pipeline breaks the task into _____ main steps.
Match each step in a warehouse robot pipeline to its main purpose.
Put the driving pipeline steps in the order of a simple perception-to-planning system.
Why does a modular self-driving system often need less labeled data for each module than a single end-to-end model?
A self-driving pipeline that separates perception, planning, and control gives the learning system an explicit breakdown of the driving task.
The third step in the drone navigation pipeline, after spotting buildings and other drones, is to _____ for the drone.
Match each concept to its meaning in a warehouse robot navigation pipeline.
Order the reasoning steps a developer follows when deciding whether to use a perception-and-planning pipeline for a self-driving car, from first to last.
Why breaking a robot navigation task into subtasks can reduce the amount of training data needed.
Choosing a design for an indoor delivery robot with little data.
Relative learning difficulty of modular driving tasks versus end-to-end control