We are announcing the general availability of Applied Intuition Copilot, a generative AI chatbot that provides users of Applied Intuition’s development platform proactive assistance in a natural language interface, empowering them to develop advanced driver-assistance systems (ADAS) and automated driving (AD) software in the most intuitive and efficient way. Applied Intuition Copilot leverages large language models (LLMs), retrieval-augmented generation (RAG), and other specialized models like diffusion and perception foundation models to provide context-aware answers and accelerate critical tasks like scenario creation and test analytics. For example, using Applied Intuition Copilot, development teams can create simulation scenarios up to 40 times faster.
ADAS and AD Development Tools
ADAS and AD systems are a critical point of differentiation for consumer vehicles and are becoming increasingly vital for commercial vehicles, whether for trucking, construction, mining, defense, or other industries. This poses a challenge to historically hardware-focused original equipment manufacturers (OEMs) and Tier 1 suppliers, as ADAS and AD systems are safety-critical, complex systems that rely heavily on both vehicle hardware and software. Handling this complexity requires OEMs and suppliers to reinvent their development workflows, which in turn often require new development toolchains.
Applied Intuition has spent over half a decade building its toolchain to accelerate the safe development of ADAS and AD systems, with its tools now supporting 18 of the 20 top global automotive OEMs and a number of enterprises and startups across other industries. These tools assist engineering teams from the early stages of system definition and development through verification, validation, and homologation, and are used by developers who write software for everything from perception to controls.
While our ADAS and AD development platform and analogous tools provided by other suppliers have improved system safety and saved developers countless hours to date, many teams across the automotive, autonomy, and robotics industries still struggle to use their tools effectively. Tools can be complex to learn, especially for non-technical users, and can require hours of manual work to complete critical tasks like scenario creation and log triage. These challenges dissuade teams from adopting new workflows and reduce the impact tools can have on program velocity.
Introducing Applied Intuition Copilot
We built Applied Intuition Copilot to tackle the usability and productivity challenges our customers have historically faced when using our ADAS and AD development platform or toolchains provided by other vendors. Copilot leverages generative AI to provide the following benefits to users of Applied Intuition Object Sim, Sensor Sim, Validation Toolset, Data Explorer, and a number of other products:
Maximize productivity with intuitive natural language interfaces
Applied Intuition Copilot provides intuitive natural language interfaces that empower developers to control Applied Intuition’s tools in a conversational manner rather than with code or powerful but sometimes complex graphical user interfaces (GUIs). This simplifies the onboarding process for all users as they can simply ask a question or provide a goal and have it answered or accomplished. It also makes the toolchain accessible to non-technical users who previously struggled to interact with code or domain-specific languages like SQL and YAML.
Accelerate development with task automation
Beyond providing the ability to control tools with natural language, Applied Intuition Copilot also automates time-intensive tasks so that engineers can get time back to focus on other activities. Generative models automate scenario creation, map creation, log tagging and triage, and a number of other common tasks. For example, teams can create scenarios up to 40x faster using Applied Intuition Copilot.
Enhance system safety with proactively generated insights
Applied Intuition Copilot goes beyond simply reacting to user commands and looks to proactively surface insights into system performance or actions that could improve development productivity. For example, Applied Intuition Copilot can automatically analyze drive logs to identify objects not captured in the current operational design domain (ODD) taxonomy and process simulation test data to identify coverage gaps or common failure modes and suggest new simulation scenarios that address them. Ultimately, Applied Intuition Copilot looks to augment the productivity of ADAS and AD development teams and ensure they are building safe systems.
