This Special Issue addresses the potential of the “Cloud” (Internet) to enhance automation and robotics for manufacturing, healthcare, transportation, logistics, security, agriculture, and many related industries by improving performance in at least five ways: 1) Big Data: indexing a global library of maps and object data; 2) Cloud Computing: parallel grid computing on demand for automation; 3) Open-Source/Open-Access: humans sharing code, data, algorithms and hardware designs; 4) System Learning: machines sharing parameters, control policies and outcomes; and 5) Crowdsourcing/call centers: offline and ondemand human guidance for evaluation, learning and error recovery.
- Scalable parallelization: How can parallel grid-based computing on demand change the current paradigm in automation science? How can parallelization schemes scale with the size of automation infrastructure?
- Effective load balancing: Balancing operations between local and cloud computation. Where should we compute for sensing, planning and actuation?
- Knowledge bases and representations: Reuse and interoperability of databases. How should online knowledge bases be shared and grow?
- Collective learning: How can an automation infrastructure search for relevant data? How can robots share and learn from experienced outcomes?
- Infrastructure/Platform or Software as a Service: To what extent can existing cloud technologies be adapted for automation and robotics? What algorithmic or technical advances are needed to allow systems to use the powerful computational, storage, and network infrastructure of data centers?
- Internet of Things: What advances complement the IoT’s sensor technologies with a physical layer for actuation? As sensors are finding their way into more Fault diagnosis and prediction for the whole production lines Prognosis and heath management for the whole production line
- Integrated and collaborative fault tolerant control
- Big Data: Data, collected and/or disseminated over large, accessible networks can enable decisions for classification problems or reveal patterns.
- Wireless communication, security and privacy issues: How can cloud-based automation be robust to latency, bandwidth limits, network failures and attacks?
- System architectures: What architectures optimize trade-offs between content aggregation and caching vs. accessibility and scalability vs. response time for automation and robotics applications?
- Open-source, open-access infrastructures: Algorithms and interfaces that introduce human feedback in automation through crowdsourcing. How can human guidance be used for evaluation, learning, and error recovery?
