From the SnowBot Pro to the Google car, self-driving robots & vehicles share many of the same core technologies

Unless you’ve been living on the moon, you’ve heard about one of the biggest innovations in the tech world: self-driving everything. Google’s driverless car may be at the leading edge, but there are thousands of self-driving robots and vehicles being developed by engineering teams inside the biggest automakers, equipment manufacturers, and young entrepreneurial tech companies.

Here at Left Hand Robotics, we’re building our own driverless robot, although I admit on a slightly smaller scale than the Google car. GPS, LiDAR, and different types of sensors are some of the core technologies being reimagined and combined in new ways to become the foundation for self-driving robots for many applications — from cars and manufacturing to industrial and outdoor tasks. We’ve incorporated these core technologies, along with many others, to develop the SnowBot Pro — our cloud-connected, sidewalk-sized self-driving robot that clears snow automatically.

Core technologies inside self-driving vehicles and robots


LiDAR stands for Light Detection and Ranging, and it measures distance by illuminating objects with a pulsed laser light then measuring the reflected pulses to calculate distance. You’ve probably seen the Roomba, the first self-driving vacuum cleaner robot. An even more advanced vacuuming robot, the Neato Botvac, uses LiDAR to calculate distance to objects in a room so it can avoid them. LiDAR is also in Google’s self-driving car to detect objects and other vehicles on the road, so it can know it needs to adjust its speed to avoid a collision.

How SnowBot Pro uses LiDAR-
We use LiDAR to detect any obstacles in the SnowBot’s path. Algorithms running onboard the SnowBot constantly analyze the sensor’s data stream to determine if an obstacle is present. If so, it immediately halts the robot and then reports to the Robot Operations Center (ROC) and to any local operators (via our mobile app) in real-time. Operators at the ROC and on local mobile devices can review the data and then instruct the robot on what to do next.

Ultrasonic and Thermal Sensors

Ultrasonic sensors are already in some technologically advanced cars on the road today, like those with automatic Reverse Parking Assist technology that use reflected sound waves to let a car park itself. Thermal sensors are able to detect objects like cars, people and animals by detecting the heat radiation emitted by an object and comparing it to the ambient radiation (heat) around the object.

How SnowBot Pro uses ultrasonic and thermal sensors-
At Left Hand Robotics, we have added ultrasonic sensors to measure the distance the SnowBot is from walls and stationary objects on each side of it — so it doesn’t bump into anything as it moves down the sidewalk. In addition, onboard thermal imaging sensors measure the heat signature of objects in front of the SnowBot’s brush to identify a pedestrian or animal that may be on the sidewalk. If an object with a heat signature is sensed, the SnowBot may slow down or stop until they safely pass.


Most self-driving vehicles take advantage of GPS in order to determine its location. However, GPS location tracking can be off by ten feet or more due to signal disturbances and other interference from the atmosphere. So to improve accuracy, real time GPS data is often compared with map data of a geographical area that has been previously collected and stored. For example, before a Google self-driving car is tested in a new area, a regular car is driven along all routes and creates a map of the area, taking note of things like road surface conditions, lane merges, traffic lights, crosswalks, and road signs.

How SnowBot uses GPS-
Our SnowBot follows a pre-programmed path that is stored in the Robot Operations Center and then downloaded to the SnowBot when it is ready to follow the path. To enable  this, during the off-season a person walks along all sidewalks and pathways they plan to have the SnowBot clear of snow and captures detailed location mapping data with our GPS Path Collection Tool – with accuracy down to the inch. This captured data is then uploaded to the ROC where it’s transformed into a Path Program that the SnowBot downloads when it is ready to start clearing snow.

Other Technologies Inside SnowBot

I’ve touched on just a few of the technologies that are at the core of many self-driving innovations. The SnowBot Pro also has an embedded computer, cellular connectivity to our cloud-based Robot Operations Center (ROC), a smartphone app to control and monitor the robot, and multiple accelerometers, gyroscopes and magnetometers for precise navigation. Our customers quickly  realize the value of SnowBot’s unique ability to capture timely snow removal data: front and rear-facing cameras take before and after photos, outside temperature is recorded, and timestamped Job Reports are created and stored automatically.

This upcoming winter season, we have many Pilot Customers across North America trialing SnowBot Pro in the field, and we can’t wait to start sharing updates and posting field videos to our blog once the snow starts to fall. Production of SnowBot Pro will be limited for the 2018-19 snow season, so check out our Priority Reservations page to learn more and get your name on the list.



About Terry Olkin

Terry is the co-founder and CEO of Left Hand Robotics. His 30-year career spans senior leadership roles at Oracle, the founding of multiple startups, and he is an inventor on over 21 patents. Prior to Left Hand Robotics, Terry was a Fellow at the public company Workday, which in 2015 acquired GridCraft, a Boulder-based software startup he co-founded. He mentors youth robotics teams competing in national competitions around the country and is the president of the nonprofit GEAR Alliance.

View All Posts