Robotics And Autonomous Systems — small scale stack.
For each requirement below, pick the option that fits your build — recommended first, then free and cheaper alternatives — or skip what your project doesn't need. Tap the info icon next to any requirement to see why it matters.
Edge Compute Fleet Management (Jetson / Greengrass)
Where your code actually runs and serves requests. Picking the right host affects speed, scaling, and how much ops work you do.Pricing & free-tier limitsFree: Greengrass free tier 1yr (100 devices), Kanto OSS free / Paid: Greengrass $0.16/device/mo + $0.00022/Lambda sec | Jetson Orin Nano $499 one-time, AGX Orin $1999 | Balena free 10 devices, then $15/device/mo billed annually | Overage: Lambda@Edge $0.0007/vCPU-sec
Device Management / IoT Core
A piece of your stack you may or may not need, depending on scope. Pick the option that fits — or skip it if your project doesn’t require this capability yet.Pricing & free-tier limitsFree: AWS IoT Core 12mo free tier 250k msgs/mo + 500 shadow ops/mo, ThingsBoard CE free self-host / Paid: AWS IoT $1.00/million msgs after, $1.08/million shadow ops | Azure IoT Hub free 8k msgs/day, S1 $25/hub/mo 400k msgs/day, S2 $250/mo 6M msgs | HiveMQ free 100 devices/10GB, paid $250/mo 1k devices
Real-time Messaging / MQTT Broker for Robot Control
A piece of your stack you may or may not need, depending on scope. Pick the option that fits — or skip it if your project doesn’t require this capability yet.Pricing & free-tier limitsFree: Mosquitto OSS $0 + EC2 $6/mo, EMQX OSS free / EMQX Cloud free tier 5M session minutes + 1GB traffic / Paid: EMQX Cloud Standard $0.80/hr per 1k conns + $0.15/GB traffic, Enterprise $1.8/hr | VerneMQ OSS free, managed $300/mo cluster | NATS JetStream free self-host, Synadia Cloud $20/mo 100GB
Time-Series DB for Sensor Data (IMU, LiDAR, Joint States)
A piece of your stack you may or may not need, depending on scope. Pick the option that fits — or skip it if your project doesn’t require this capability yet.Pricing & free-tier limitsFree: InfluxDB OSS free unlimited self-host, InfluxDB Cloud free 30GB storage + 300M series cardinality, 300GB data out | Timescale Community free | Paid: Timescale Cloud $55/mo 500GB storage/50GB backup, $0.12/GB ingest | InfluxDB Cloud Serverless $0.002/MB write, $0.01/GB query, $0.25/GB-hr storage | QuestDB OSS free, Cloud $70/mo 10GB
Simulation Environment (Isaac Sim, Gazebo on GPU)
Stores API keys and credentials safely, separate from your code. Leaking secrets is one of the most common breaches.Pricing & free-tier limitsFree: Gazebo OSS free, Isaac Sim free for individuals | Paid: Isaac Sim Enterprise $1,500/seat/yr + OVX server $9k/yr | AWS g5.xlarge $1.006/hr on-demand, spot $0.35/hr, 4x A100 g5.48xl $16.28/hr, A100 $1.5-2.1/hr spot $0.9/hr, H100 $4.2/hr spot $2.4/hr | GCP L4 $0.70/hr
GPU Training Cluster for Robotics Models (RL, VLA, Diffusion Policy)
Powers AI features — model access, embeddings, and inference. Costs scale with usage, so watch the meter.Pricing & free-tier limitsFree: Colab T4 free 12hr sess, Kaggle 30hr P100/wk + 2x T4 x2 free | Paid: RunPod A100 $1.14/hr Secure, community $0.85/hr, H100 $2.29/hr | Lambda A100 40GB $1.10/hr, H100 $2.49/hr (1-yr reserved $1.79) | CoreWeave H100 80GB $3.49/hr | AWS p5.48xl $98.32/hr (8x H100) ~$12.29/GPU/hr, A100 p4d.24xl $32.77/hr | Modal H100 $3.5/hr + $0.0006/CPU-sec free 30 hrs/mo $30 credits
Data Lake for Sensor Logs / ROS Bag Storage (Petabyte)
Where you keep files users upload or you serve — images, videos, documents — and how fast they reach visitors around the world.Pricing & free-tier limitsFree: S3 free tier 5GB-std 12mo + MinIO OSS free | R2 free 10GB storage + 10M Class A ops, zero egress | Backblaze B2 free 10GB | Paid: S3 Std $0.023/GB/mo, IA $0.0125, Glacier Instant $0.004, Deep $0.00099 | R2 $0.015/GB/mo + $4.50/M Class A, $0.36/M Class B | B2 $0.006/GB/mo + $0.01/GB egress first 1GB/day free | S3 Select $0.002/GB scanned
Video Storage for Robot Cameras (Fleet DVR)
Where you keep files users upload or you serve — images, videos, documents — and how fast they reach visitors around the world.Pricing & free-tier limitsFree: MinIO OSS free, KVS free ingested 10k mins/mo first year, S3 5GB | Paid: S3 $0.023/GB/mo, Kinesis Video $0.0085/GB-ingest + $0.0119/GB storage/mo + H264 transcoding $0.016/stream-hr | Wasabi $6.99/TB/mo flat no egress | Cloudflare Stream $5/1000min stored + $1/1000min delivered | Mux $0.01/min stored
Container Orchestration (Cloud + Edge K3s)
Where you keep files users upload or you serve — images, videos, documents — and how fast they reach visitors around the world.Pricing & free-tier limitsFree: K3s OSS $0, k3d free, EKS free tier 1 cluster no mgmt fee first 3mo then $0.10/hr | Paid: EKS $0.10/hr/cluster = $73/mo + EC2 worker m6i.large $0.096/hr, EKS Anywhere $0 + support $300/mo | DOKS free control plane + $12/mo droplet, Civo $10/mo per node + $6 TLB | Rancher free
Object Storage for Models & Artifacts
Where you keep files users upload or you serve — images, videos, documents — and how fast they reach visitors around the world.Pricing & free-tier limitsFree: HF Hub 100GB free private, S3 5GB free 12mo, R2 10GB free | Paid: S3 $0.023/GB/mo, R2 $0.015/GB/mo zero egress, B2 $0.006/GB/mo $0.01/GB download | LakeFS OSS free, Cloud $100/mo 100GB + $0.10/GB | Overages: S3 PUT $0.005/1k, GET $0.0004/1k
Secrets Management for Robot Credentials (Certs, Tokens)
A piece of your stack you may or may not need, depending on scope. Pick the option that fits — or skip it if your project doesn’t require this capability yet.Pricing & free-tier limitsFree: Vault OSS free self-host, Doppler free 5 users + 200 secrets | Paid: AWS Secrets $0.40/secret/mo + $0.05/10k API calls, first 30days free | HCP Vault $0.03/hr + $0.50/secret/mo Dedicated $1.5/hr cluster | Doppler Team $8/user/mo, Enterprise $23/user | Infisical Cloud free 500 secrets, $30/mo team
OTA Update Service for Robots (Mender, JFrog)
A piece of your stack you may or may not need, depending on scope. Pick the option that fits — or skip it if your project doesn’t require this capability yet.Pricing & free-tier limitsFree: Mender OSS free unlimited self-host (10 devices hosted free), RAUC free, AWS IoT Jobs free tier included | Paid: Mender hosted $135/mo Starter 15 devices, Professional $430/mo 100 devices, $2.50/device/mo over 250, Enterprise $1.2k/mo 500 dev | JFrog Connect free 10 dev, $19/mo 50 dev, Pro $99/mo 250 dev | Balena $15/dev/mo >10 | AWS IoT Jobs $0.002/job execution
Digital Twin Platform
A piece of your stack you may or may not need, depending on scope. Pick the option that fits — or skip it if your project doesn’t require this capability yet.Pricing & free-tier limitsFree: Eclipse Ditto OSS free, TwinMaker first 3 scenes free | Paid: TwinMaker entity query $0.04/1000, $0.46/1000 data access, rendering $0.30/hr, S3 + Lambda cost | MindSphere $250/mo Starter, $1500/mo Pro | Ditto managed $70/mo small
GPU Inference Hosting (Triton, ONNX Runtime Server)
Where your code actually runs and serves requests. Picking the right host affects speed, scaling, and how much ops work you do.Pricing & free-tier limitsFree: Triton OSS free, Modal $30 free credits/mo + free 30 GPU-hr, Replicate free $5 credits | Paid: Modal A10G $0.45/hr + $0.0003/req, H100 $2.9/hr | Baseten $0.20-1.50/hr + $0.0002/inference (A10G $0.32/hr) | Replicate L40S $0.000725/sec ~$2.61/hr | Anyscale A10G $1.2/hr, H100 $4.5/hr first 1M tok free | Triton on EKS $0.10/hr + GPU EC2 $1.5/hr A10G
Model Registry & Versioning (ROS Models + AI)
Powers AI features — model access, embeddings, and inference. Costs scale with usage, so watch the meter.Pricing & free-tier limitsFree: MLflow OSS free, DVC free, HF Hub 100GB free | W&B Free 100GB tracked, personal $0 | Paid: W&B Team $25/user/mo 100GB included overage $0.12/GB, Enterprise $75/user | Databricks DBU $0.55/DBU + $0.10/registry-hr + storage $0.023/GB | JFrog ML Pro $30/mo | MLflow managed Databricks $0 extra
Monitoring for Fleet (Metrics, Uptime, Telemetry)
Tells you when things break and why. You cannot fix what you cannot see — this is how you keep downtime short.Pricing & free-tier limitsFree: Grafana Cloud free 10k series + 50GB logs + 50GB traces + 14d retention, Prometheus OSS free | Paid: Grafana Cloud Pro $29/mo 15k series + $8/1000 series + $0.50/GB logs, Enterprise $199/mo | Datadog IoT $5/device/mo 10 devices min + logs $0.10/GB ingest + infra $15/host/mo | New Relic free 100GB/mo, Pro $0.30/GB after
Centralized Logging (ROS Logs, Fleet Logs)
A piece of your stack you may or may not need, depending on scope. Pick the option that fits — or skip it if your project doesn’t require this capability yet.Pricing & free-tier limitsFree: Loki OSS free unlimited self-host, Grafana Cloud 50GB logs free, Better Stack free 1GB/mo | Paid: Grafana Loki $0.50/GB ingest + $0.06/GB-mo storage, Better Stack $24/mo 10GB 7d retain $0.02/GB over, Datadog $0.10/GB ingest + $1.70/million log events + $0.10/GB-scan | Elastic Cloud $95/mo 2GB RAM ingest 50GB storage
Relational DB for Fleet Metadata, Missions, Users
A piece of your stack you may or may not need, depending on scope. Pick the option that fits — or skip it if your project doesn’t require this capability yet.Pricing & free-tier limitsFree: Supabase free 500MB DB + 50k MAU, Neon free 0.5GB + 10GB transfer 100hr compute, Local PG free | Paid: Neon Launch $19/mo 10GB + $5/10GB extra + $10/100 compute-hr | Supabase Pro $25/mo 8GB + $0.125/GB extra, $0.00365/hr compute | RDS db.t3.micro $0.017/hr ~$12/mo + 20GB $0.115/GB-mo, Multi-AZ $0.034/hr, Aurora Serverless v2 $0.12/ACU-hr
VPN / Secure Connectivity for Field Robots
A piece of your stack you may or may not need, depending on scope. Pick the option that fits — or skip it if your project doesn’t require this capability yet.Pricing & free-tier limitsFree: Tailscale free 100 devices + 3 users ACL unlimited, ZeroTier free 25 nodes + 1 admin, WireGuard $0 + $5/mo VPS | Paid: Tailscale Starter $6/user/mo + free 3 users, Teams $18/user/mo + $120/mo min, AWS Secure Tunneling $0.016/tunnel-min 60min free tier? $1.01/hr tunnel, data $0.01/GB | ZeroTier Pro $5/member/mo + $100/mo org
How this small scale robotics and autonomous systems checklist works.
Each requirement below is something a small scale robotics and autonomous systems build typically needs. Pick one of the four researched options — recommended, free, cheaper or paid — add your own with "Other", or skip the requirement if your project doesn't need it. Nothing is mandatory; the plan on the right tracks what you've decided so nothing gets forgotten.
Your picks are saved in this browser automatically, so you can come back anytime. Options are researched per build level and refreshed as vendors change their plans — always verify details on the provider's page before committing.